[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Issue
Volume 14, January-1
Previous Issue
Volume 13, December-1
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
8.3
4.2
Journals
Remote Sensing
Volume 13
Issue 24
remotesensing-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Remote Sens., Volume 13, Issue 24 (December-2 2021) – 227 articles

Cover Story (view full-size image): The availability of huge datasets of satellite remote sensing measurements has fostered the development of fast, efficient retrieval codes. Deep learning techniques were recently applied to satellite retrievals. Forward models are a fundamental part of retrieval code development and mission design. However, the application of deep learning techniques to radiative transfer simulations is still underexplored. Here, deep learning techniques are applied to the design of the retrieval chain of an upcoming satellite mission, LSTM, a candidate for Sentinel 9: they are used to generate spectral features and analyze simulated observations. The performance of deep learning algorithms shows promising results for the production of both simulated spectra and parameter retrievals, one of the main advances being the reduction in computational costs.View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
17 pages, 16605 KiB  
Article
Toward a Simple and Generic Approach for Identifying Multi-Year Cotton Cropping Patterns Using Landsat and Sentinel-2 Time Series
by Qiqi Li, Guilin Liu and Weijia Chen
Remote Sens. 2021, 13(24), 5183; https://doi.org/10.3390/rs13245183 - 20 Dec 2021
Cited by 8 | Viewed by 3841
Abstract
The sustainable development goals of the United Nations, as well as the era of pandemics have introduced serious challenges for agricultural production and management. Precise management of agricultural practices based on satellite-borne remote sensing has been considered an effective means for monitoring cropping [...] Read more.
The sustainable development goals of the United Nations, as well as the era of pandemics have introduced serious challenges for agricultural production and management. Precise management of agricultural practices based on satellite-borne remote sensing has been considered an effective means for monitoring cropping patterns and crop-farming patterns. Therefore, we proposed a simple and generic approach to identify multi-year cotton-cropping patterns based on time series of Landsat and Sentinel-2 images, with few ground samples that covered many years, a simple classification algorithm, and had a high classification accuracy. In this approach, we extended the size of training samples using active learning, and we employed a random forest algorithm to extract multi-year cotton planting patterns based on dense time series of Landsat and Sentinel-2 data from 2014 to 2018. We created annual crop cultivation maps based on training samples with an accuracy greater than 95.69%. The accuracy of multi-year cotton cropping patterns was 96.93%. The proposed approach was effective and robust in identifying multi-year cropping patterns, and it could be applied in other regions. Full article
(This article belongs to the Special Issue Remote Sensing Applications in Agricultural Ecosystems)
►▼ Show Figures

Figure 1

Figure 1
<p>Location of the study area in China.</p> Full article ">Figure 2
<p>Available Landsat and Sentinel-2 time series that covered the study area from 2014 to 2018.</p> Full article ">Figure 3
<p>Number of samples for each annual year, in which 30% are training samples and 70% are verification samples.</p> Full article ">Figure 4
<p>Flow chart of this study.</p> Full article ">Figure 5
<p>Temporal <span class="html-italic">EVI</span> of different land cover types in 2014.</p> Full article ">Figure 6
<p>(<b>a</b>) Five-year temporal <span class="html-italic">EVI</span> of cotton-rice rotation. (<b>b</b>) Five-year temporal <span class="html-italic">EVI</span> of cotton succession (rotation, reclamation, fallow, and abandonment).</p> Full article ">Figure 7
<p>Annual cotton classification maps over the 5-year period from 2014 to 2018.</p> Full article ">Figure 7 Cont.
<p>Annual cotton classification maps over the 5-year period from 2014 to 2018.</p> Full article ">Figure 8
<p>Multi-year cotton-cropping pattern between 2014 and 2018.</p> Full article ">Figure 9
<p>Area of multi-year cotton-cropping patterns from 2014 to 2018; (<b>a</b>) pixel-based multi-year cotton cropping pattern, (<b>b</b>) GIS-driven multi-year cotton cropping pattern.</p> Full article ">Figure 10
<p>GIS-driven multi-year cotton-planting pattern from 2014 to 2018.</p> Full article ">Figure 11
<p>Cotton cultivation patterns at the Aksu Station; all remote sensing images are in July of each year; (<b>a</b>) multi-year cotton cropping pattern using RF method, (<b>b</b>) GIS-driven multi-year cotton cropping pattern.</p> Full article ">
22 pages, 128572 KiB  
Article
Deep Learning-Based Object Detection System for Identifying Weeds Using UAS Imagery
by Aaron Etienne, Aanis Ahmad, Varun Aggarwal and Dharmendra Saraswat
Remote Sens. 2021, 13(24), 5182; https://doi.org/10.3390/rs13245182 - 20 Dec 2021
Cited by 37 | Viewed by 6123
Abstract
Current methods of broadcast herbicide application cause a negative environmental and economic impact. Computer vision methods, specifically those related to object detection, have been reported to aid in site-specific weed management procedures for targeted herbicide application within a field. However, a major challenge [...] Read more.
Current methods of broadcast herbicide application cause a negative environmental and economic impact. Computer vision methods, specifically those related to object detection, have been reported to aid in site-specific weed management procedures for targeted herbicide application within a field. However, a major challenge to developing a weed detection system is the requirement for a properly annotated database to differentiate between weeds and crops under field conditions. This research involved creating an annotated database of 374 red, green, and blue (RGB) color images organized into monocot and dicot weed classes. The images were acquired from corn and soybean research plots located in north-central Indiana using an unmanned aerial system (UAS) flown at 30 and 10 m heights above ground level (AGL). A total of 25,560 individual weed instances were manually annotated. The annotated database consisted of four different subsets (Training Image Sets 1–4) to train the You Only Look Once version 3 (YOLOv3) deep learning model for five separate experiments. The best results were observed with Training Image Set 4, consisting of images acquired at 10 m AGL. For monocot and dicot weeds, respectively, an average precision (AP) score of 91.48 % and 86.13% was observed at a 25% IoU threshold (AP @ T = 0.25), as well as 63.37% and 45.13% at a 50% IoU threshold (AP @ T = 0.5). This research has demonstrated a need to develop large, annotated weed databases to evaluate deep learning models for weed identification under field conditions. It also affirms the findings of other limited research studies utilizing object detection for weed identification under field conditions. Full article
(This article belongs to the Special Issue Advances of Remote Sensing in Precision Agriculture)
►▼ Show Figures

Figure 1

Figure 1
<p>Data acquisition dates for 2018 and 2019. Each date also denotes where UAS flights were performed to collect data.</p> Full article ">Figure 2
<p>Corn and soybean growth stages (adapted from [<a href="#B52-remotesensing-13-05182" class="html-bibr">52</a>,<a href="#B53-remotesensing-13-05182" class="html-bibr">53</a>]).</p> Full article ">Figure 3
<p>YOLOv3 network architecture—detection at three different scales are combined for the final detection (Figure adapted from [<a href="#B55-remotesensing-13-05182" class="html-bibr">55</a>]).</p> Full article ">Figure 4
<p>LabelImg manual labeling process and output.</p> Full article ">Figure 5
<p>Training image set creation flowchart. All training image sets are trained with YOLOv3.</p> Full article ">Figure 6
<p>Illustration of the IoU on the dicot dataset images.</p> Full article ">Figure 7
<p>Monocot and dicot detection results from the YOLOv3 network training on Image Set 1.</p> Full article ">Figure 8
<p>Map and loss scores over each training iteration from YOLOv3 network training on Image Set 1.</p> Full article ">Figure 9
<p>Monocot weed detection test set result from the YOLOv3 training on Image Set 2.</p> Full article ">Figure 10
<p>Dicot detection test set result from the YOLOv3 training on Image Set 3.</p> Full article ">Figure 11
<p>Dicot detection result from the YOLOv3 network training on Training Image Set 4+, using the Gilbreth cluster.</p> Full article ">
15 pages, 2654 KiB  
Article
Using CYGNSS Data to Map Flood Inundation during the 2021 Extreme Precipitation in Henan Province, China
by Shuangcheng Zhang, Zhongmin Ma, Zhenhong Li, Pengfei Zhang, Qi Liu, Yang Nan, Jingjiang Zhang, Shengwei Hu, Yuxuan Feng and Hebin Zhao
Remote Sens. 2021, 13(24), 5181; https://doi.org/10.3390/rs13245181 - 20 Dec 2021
Cited by 46 | Viewed by 4896
Abstract
On 20 July 2021, parts of China’s Henan Province received the highest precipitation levels ever recorded in the region. Floods caused by heavy rainfall resulted in hundreds of casualties and tens of billions of dollars’ worth of property loss. Due to the highly [...] Read more.
On 20 July 2021, parts of China’s Henan Province received the highest precipitation levels ever recorded in the region. Floods caused by heavy rainfall resulted in hundreds of casualties and tens of billions of dollars’ worth of property loss. Due to the highly dynamic nature of flood disasters, rapid and timely spatial monitoring is conducive for early disaster prevention, mid-term disaster relief, and post-disaster reconstruction. However, existing remote sensing satellites cannot provide high-resolution flood monitoring results. Seeing as spaceborne global navigation satellite system-reflectometry (GNSS-R) can observe the Earth’s surface with high temporal and spatial resolutions, it is expected to provide a new solution to the problem of flood hazards. Here, using the Cyclone Global Navigation Satellite System (CYGNSS) L1 data, we first counted various signal-to-noise ratios and the corresponding reflectivity to surface features in Henan Province. Subsequently, we analyzed changes in the delay-Doppler map of CYGNSS when the observed area was submerged and not submerged. Finally, we determined the submerged area affected by extreme precipitation using the threshold detection method. The results demonstrated that the flood range retrieved by CYGNSS agreed with that retrieved by the Soil Moisture Active Passive (SMAP) mission and the precipitation data retrieved and measured by the Global Precipitation Measurement mission and meteorological stations. Compared with the SMAP results, those obtained by CYGNSS have a higher spatial resolution and can monitor changes in the areas affected by the floods over a shorter period. Full article
(This article belongs to the Special Issue Recent Advances in GNSS Reflectometry)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Digital Elevation Model (DEM) in Henan Province; (<b>b</b>) precipitation data measured by some meteorological stations; (<b>c</b>) precipitation in Henan Province on 20 July (Global Precipitation Measurement (GPM)-derived); (<b>d</b>) precipitation in Henan Province on 21 July (GPM-derived).</p> Full article ">Figure 2
<p>(<b>a</b>) Land use and cover in Henan Province. Note: There are two main rice fields in Henan Province, one in the south (red box) and the other in the north (black box). The planting cycle of the southern rice-producing area is from early May to early September, and the planting cycle of the northern rice-producing area is from mid-July to mid-October. Therefore, during this research period, the rice-producing area in the north was actually planted as upland crops, which can be considered as upland. (<b>b</b>) Cyclone Global Navigation Satellite System (CYGNSS) surface reflectivity (SR) (scattered form) from 1 June to 5 June.</p> Full article ">Figure 3
<p>Delay-Doppler map (DDM) changes before and after floods in the same area. (<b>a</b>) 5 June (before the flood); (<b>b</b>) 23 July (flooding); (<b>c</b>) 13 August (after the flood).</p> Full article ">Figure 4
<p>Soil Moisture Active Passive (SMAP) soil moisture changes before and after floods in Henan Province and the results of SMAP inversion of flood areas. (<b>a</b>) SMAP soil moisture on 19 June; (<b>b</b>) SMAP soil moisture on 26 June; (<b>c</b>) SMAP soil moisture on 20 July; (<b>d</b>) SMAP soil moisture on 23 July; (<b>e</b>) SMAP inversion results of floods in the first stage of heavy rainfall (16–20 July); (<b>f</b>) SMAP inversion results of floods in the second stage of heavy rainfall (21–25 July).</p> Full article ">Figure 5
<p>Gridded Cyclone Global Navigation Satellite System surface reflectivity (CYGNSS SR) before and after extreme precipitation in Henan Province. (<b>a</b>) Gridded SR before the occurrence of heavy rainfall (3 × 3 km); (<b>b</b>) Gridded SR in the first stage of heavy rainfall (6 × 6 km); (<b>c</b>) Gridded SR in the second stage of heavy rainfall (6 × 6 km).</p> Full article ">Figure 6
<p>The distribution map of surface inundation before and after heavy rainfall in Henan Province obtained by Cyclone Global Navigation Satellite System surface reflectivity (CYGNSS SR). (<b>a</b>) Distribution of surface inundation before the occurrence of extreme precipitation (1 June 2021–30 June 2021); (<b>b</b>) Distribution of floods in the first stage of extreme precipitation (16 July 2021–20 July 2021); (<b>c</b>) Flood distribution in the second stage of extreme precipitation (21 July 2021–25 July 2021). The red pentagram represents the locations of Zhengzhou, Xinxiang, Hebi, and Luoyang. The corresponding relationship is shown in <a href="#remotesensing-13-05181-f001" class="html-fig">Figure 1</a>.</p> Full article ">Figure 7
<p>Changes in precipitation, soil moisture, and reflectivity near (34° N, 114° E) from 1 June to 31 August 2021. Rainfall (black bar graph), soil moisture (red scatter graph), and surface reflectivity (SR) (blue dotted line graph). r = 0.7.</p> Full article ">
14 pages, 7173 KiB  
Technical Note
Tropospheric Attenuation in GeoSurf Satellite Constellations
by Emilio Matricciani, Carlo Riva and Lorenzo Luini
Remote Sens. 2021, 13(24), 5180; https://doi.org/10.3390/rs13245180 - 20 Dec 2021
Cited by 5 | Viewed by 2381
Abstract
In GeoSurf satellite constellations, any transmitter/receiver, wherever it is located, is linked to a satellite with zenith paths. We have studied the tropospheric attenuation predicted for some reference sites (Canberra, Holmdel, Pasadena, Robledo, and Spino d’Adda), which also set the meridian along which [...] Read more.
In GeoSurf satellite constellations, any transmitter/receiver, wherever it is located, is linked to a satellite with zenith paths. We have studied the tropospheric attenuation predicted for some reference sites (Canberra, Holmdel, Pasadena, Robledo, and Spino d’Adda), which also set the meridian along which we have considered sites with latitudes ranging between 60° N and 60° S. At the annual probability of 1% of an average year, in the latitude between 30° N and 30° S, there are no significant differences between GEO slant paths and GeoSurf zenith paths. On the contrary, at 0.1% and 0.01% annual probabilities, large differences are found for latitudes greater than 30° N or 30° S. For comparing the tropospheric attenuation in GeoSurf paths with that expected in LEO highly variable slant paths, we have considered, as reference, a LEO satellite constellation orbiting in circular at 817 km. GeoSurf zenith paths “gain” several dBs compared to LEO slant paths. The more static total clear-sky attenuation (water vapor, oxygen, and clouds) in both GEO and LEO slant paths shows larger values than GeoSurf zenith paths. Both for rain and clear-sky attenuations, Northern and Southern Hemispheres show significant differences. Full article
(This article belongs to the Special Issue Multiple Access Edge Computing in Non-Terrestial and Terrestrial Internet of Things Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Annual probability distributions (%) of rain attenuation exceeded in the slant path at Spino d’Adda to a geostationary satellite (elevation angle 37.7°) on the meridian, at 39.6 GHz and a circular polarization, calculated by the indicated prediction models, using, as input, the rain rate probability distribution recorded locally, with rain rate expressed in millimeters per hour, in the years 1993−2002. The solid red line refers to using the full SST [<a href="#B17-remotesensing-13-05180" class="html-bibr">17</a>], whose input is the full set of rain rate time series recorded in the same period of time, whose probability distribution is used as input to the models.</p> Full article ">Figure 2
<p>Annual probability distributions (%) of rain attenuation exceeded in the slant paths directed along the local meridian to a Geostationary satellite at 39.6 GHz and a circular polarization in: (<b>a</b>) Robledo (elevation angle 42.8°): (<b>b</b>) Holmdel (elevation angle 43.3°), (<b>c</b>) Pasadena (elevation angle 50.3°), and (<b>d</b>) Canberra (elevation angle 49.0), predicted by using the ITU−R [<a href="#B28-remotesensing-13-05180" class="html-bibr">28</a>,<a href="#B29-remotesensing-13-05180" class="html-bibr">29</a>] rain rate probability distributions.</p> Full article ">Figure 3
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math> (dB), calculated at fixed annual probabilities 1% (<b>a</b>) 0.1% (<b>b</b>), and 0.01%, (<b>c</b>) between the rain attenuation exceeded in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math> (dB), and the rain attenuation exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math> (dB), i.e., at the zenith in Spino d’Adda; 39.6 GHz, circular polarization.</p> Full article ">Figure 4
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at fixed annual probabilities 1% (<b>a</b>), 0.1% (<b>b</b>), and 0.01% (<b>c</b>), between the rain attenuation exceeded in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and the rain attenuation exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, i.e., at the zenith in Robledo; 39.6 GHz, circular polarization.</p> Full article ">Figure 5
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at fixed annual probabilities 1% (<b>a</b>), 0.1% (<b>b</b>), and 0.01% (<b>c</b>), between the rain attenuation exceeded in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and the rain attenuation exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, i.e., at the zenith in Holmdel; 39.6 GHz, circular polarization.</p> Full article ">Figure 6
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at fixed annual probabilities 1% (<b>a</b>), 0.1% (<b>b</b>), and 0.01% (<b>c</b>), between the rain attenuation exceeded in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and the rain attenuation exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, i.e., at the zenith in Pasadena; 39.6 GHz, circular polarization.</p> Full article ">Figure 7
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at fixed annual probabilities 1% (<b>a</b>), 0.1% (<b>b</b>), and 0.01% (<b>c</b>), between the rain attenuation exceed in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and the rain attenuation exceed in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, i.e., at the zenith in Pasadena; 39.6 GHz, circular polarization.</p> Full article ">Figure 8
<p>Orbits and tracks on ground of Metop−C satellite. 20 October 2021; starting time: 20:07:26 UTC.</p> Full article ">Figure 9
<p>Sample visibility windows: Metop−C satellite seen from Spino d’Adda. (<b>a</b>): LEO satellite elevation angle. (<b>b</b>): LEO satellite azimuth.</p> Full article ">Figure 10
<p>Probability density of the Metop−C satellite elevation angle seen from Spino d’Adda for elevation angles larger than <math display="inline"><semantics> <mrow> <mo>~</mo> <mn>20</mn> <mo>°</mo> </mrow> </semantics></math>.</p> Full article ">Figure 11
<p>Results predicted with the Global SST model. (<b>a</b>) LEO annual probability distributions (%) of rain attenuation at the reference sites, obtained by weighing the probability distribution of rain attenuation predicted for each elevation angle larger than 20° according to the probability of the elevation angle shown in <a href="#remotesensing-13-05180-f010" class="html-fig">Figure 10</a>; 39.6 GHz, at the reference sites. (<b>b</b>) The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>L</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, vs. annual probability (%), between the rain attenuation exceed in the LEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>L</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and the rain attenuation exceed in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, i.e., at the zenith at the reference sites.</p> Full article ">Figure 12
<p>Annual probability distributions of clear-sky attenuation exceeded in the GEO paths on the local meridian at the reference sites.</p> Full article ">Figure 13
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at annual probabilities 1%, 5%, and 10%, between the clear-sky attenuation exceeded in the GEO path (elevation angle 37.7°), <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and that exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, along the meridian of Spino d’Adda (9.5°), in the latitude range 60° N to 60° S.</p> Full article ">Figure 14
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated at annual probabilities 1%, 5%, and 10%, between the clear-sky attenuation exceeded in the GEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>G</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and that exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, along the meridian of: (<b>a</b>) Robledo (GEO elevation angle 42.8°), (<b>b</b>) Holmdel (GEO elevation angle 43.3°), (<b>c</b>) Pasadena (GEO elevation angle 50.3°), and (<b>d</b>) Canberra (GEO elevation angle 49.0°), in the latitude range 60° N to 60° S.</p> Full article ">Figure 15
<p>The difference <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>L</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> <mo>−</mo> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, calculated, at equal probability, in the probability range 1% to 10% between the clear-sky attenuation exceeded in the LEO path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mrow> <mi>L</mi> <mi>E</mi> <mi>O</mi> </mrow> </msub> </mrow> </semantics></math>, and that exceeded in the GeoSurf path, <math display="inline"><semantics> <mrow> <msub> <mi>A</mi> <mi>Z</mi> </msub> </mrow> </semantics></math>, in the reference sites.</p> Full article ">
16 pages, 2326 KiB  
Article
Susceptibility of East Asian Marine Warm Clouds to Aerosols in Winter and Spring from Co-Located A-Train Satellite Observations
by Chiao-Wei Chang, Wei-Ting Chen and Yi-Chun Chen
Remote Sens. 2021, 13(24), 5179; https://doi.org/10.3390/rs13245179 - 20 Dec 2021
Cited by 1 | Viewed by 2676
Abstract
We constructed the A-Train co-located aerosol and marine warm cloud data from 2006 to 2010 winter and spring over East Asia and investigated the sensitivities of single-layer warm cloud properties to aerosols under different precipitation statuses and environmental regimes. The near-surface stability (NSS), [...] Read more.
We constructed the A-Train co-located aerosol and marine warm cloud data from 2006 to 2010 winter and spring over East Asia and investigated the sensitivities of single-layer warm cloud properties to aerosols under different precipitation statuses and environmental regimes. The near-surface stability (NSS), modulated by cold air on top of a warm surface, and the estimated inversion strength (EIS) controlled by the subsidence are critical environmental parameters affecting the marine warm cloud structure over East Asia and, thus, the aerosols–cloud interactions. Based on our analysis, precipitating clouds revealed higher cloud susceptibility to aerosols as compared to non-precipitating clouds. The cloud liquid water path (LWP) increased with aerosols for precipitating clouds, yet decreased with aerosols for non-precipitating clouds, consistent with previous studies. For precipitating clouds, the cloud LWP and albedo increased more under higher NSS as unstable air promotes more moisture flux from the ocean. Under stronger EIS, the cloud albedo response to aerosols was lower than that under weaker EIS, indicating that stronger subsidence weakens the cloud susceptibility due to more entrainment drying. Our study suggests that the critical environmental factors governing the aerosol–cloud interactions may vary for different oceanic regions, depending on the thermodynamic conditions. Full article
(This article belongs to the Section Atmospheric Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>Example of marine stratocumulus during a cold surge event over East Asia from the true color image of MOderate Resolution Imaging Spectroradiometer (MODIS) on 13 January 2010. Left: at 04:45 UTC, overpassing northern China, the Korean Peninsula, and the Yellow Sea. Right: 04:40 UTC, overpassing southern China, Taiwan, and the northwestern Pacific. (<a href="https://ladsweb.modaps.eosdis.nasa.gov/search/imageViewer/42/MYD021KM--61/2010-01-13/DB/World/3077701375" target="_blank">https://ladsweb.modaps.eosdis.nasa.gov/search/imageViewer/42/MYD021KM--61/2010-01-13/DB/World/3077701375</a>, accessed on 27 October 2021).</p> Full article ">Figure 2
<p>The processes of selecting marine warm cloud pixels over East Asia.</p> Full article ">Figure 3
<p>Spatial distribution of the marine warm cloud pixels: (<b>a</b>) All, (<b>b</b>) Non–precipitating, (<b>c</b>) Precipitating.</p> Full article ">Figure 4
<p>(<b>a</b>,<b>b</b>) The 2006–2010 DJF and MAM mean 850 hPa wind field, respectively. (<b>c</b>,<b>d</b>) The 2006–2010 DJF and MAM mean SST, respectively. These are ERA-Interim reanalysis data.</p> Full article ">Figure 5
<p>(<b>a</b>,<b>b</b>) The 2006–2010 DJF and MAM mean NSS (unit: K), respectively. (<b>c</b>,<b>d</b>) The 2006–2010 DJF and MAM mean EIS (unit: K), respectively. (<b>e</b>,<b>f</b>) The 2006–2010 DJF and MAM mean AI, respectively. NSS and EIS are derived from ECMWF-AUX. All the filled grids are averaged over the <math display="inline"><semantics> <mrow> <mn>5</mn> <mo>°</mo> <mo>×</mo> <mn>5</mn> <mo>°</mo> </mrow> </semantics></math> grid, which contains more than 100 pixels.</p> Full article ">Figure 6
<p>The boxplot of LWP (unit:gm<sup>−2</sup>), cloud fraction (unit: %), cloud top pressure (unit: hPa), cloud optical depth, and AI in each NSS bin. The numbers on the x-axis denote 10th, 30th, 50th, 70th, and 90th percentiles by sequence. (<b>a</b>,<b>c</b>,<b>e</b>,<b>g</b>,<b>i</b>) For non-precipitating clouds. (<b>b</b>,<b>d</b>,<b>f</b>,<b>h</b>,<b>j</b>) For precipitating clouds. The median of each dataset is shown in red; the lower quartile (25th percentile, or Q<sub>1</sub>) and the upper quartile (75th percentile, or Q<sub>3</sub>) are shown as the lower and upper boundary of the blue box, respectively.</p> Full article ">Figure 7
<p>(<b>a</b>,<b>c</b>,<b>e</b>,<b>g</b>,<b>i</b>) For non-precipitating clouds. (<b>b</b>,<b>d</b>,<b>f</b>,<b>h</b>,<b>j</b>) For precipitating clouds. The same as <a href="#remotesensing-13-05179-f006" class="html-fig">Figure 6</a>, but for different EIS bins. The numbers on the x-axis denote 10th, 30th, 50th, 70th, and 90th percentiles by sequence.</p> Full article ">Figure 8
<p>The cloud susceptibility to aerosol index (AI) under stable and unstable conditions. The cloud parameters were: (<b>a</b>) Cloud droplet effective radius (Re), (<b>b</b>) Cloud liquid water path (LWP), (<b>c</b>) Cloud optical depth (τ), and (<b>d</b>) Cloud albedo. The threshold value for the stable and unstable conditions was defined as NSS &lt; 25th percentile and NSS &gt; 75th percentile, respectively. Response for precipitating and non-precipitating clouds are shown in green and blue, respectively. All the filled bars are statistically significant at a 95% significance level. The error bars were calculated based on the standard error of the regression slope.</p> Full article ">Figure 9
<p>The cloud susceptibility to aerosol index (AI) under weak and strong inversion strengths. The cloud parameters were: (<b>a</b>) Cloud droplet effective radius (Re), (<b>b</b>) Cloud liquid water path (LWP), (<b>c</b>) Cloud optical depth, and (<b>d</b>) Cloud albedo. The threshold value for weak and strong inversion was defined as EIS &lt; 25th percentile and EIS &gt; 75th percentile, respectively. Response for precipitating and non-precipitating clouds are shown in green and blue, respectively. All the filled bars are statistically significant at a 95% significance level. The error bars were calculated based on the standard error of the regression slope.</p> Full article ">
16 pages, 28707 KiB  
Article
Electromagnetic Scattering of Near-Field Turbulent Wake Generated by Accelerated Propeller
by Yuxin Deng, Min Zhang, Wangqiang Jiang and Letian Wang
Remote Sens. 2021, 13(24), 5178; https://doi.org/10.3390/rs13245178 - 20 Dec 2021
Cited by 3 | Viewed by 2769
Abstract
The electromagnetic scattering study of the turbulent wake of a moving ship has important application value in target recognition and tracking. However, to date, there has been insufficient research into the electromagnetic characteristics of near-field propeller turbulence. This study presents a new procedure [...] Read more.
The electromagnetic scattering study of the turbulent wake of a moving ship has important application value in target recognition and tracking. However, to date, there has been insufficient research into the electromagnetic characteristics of near-field propeller turbulence. This study presents a new procedure for evaluating the electromagnetic scattering coefficient and imaging characteristics of turbulent wakes in the near field. By controlling the different values of the net momenta, a turbulent wake was generated using the large-eddy simulation method. The results show that the net momentum transferred to the background flow field determines the development of the turbulent wake, which explains the formation mechanism of the turbulence. Combined with the turbulent energy attenuation spectrum, the electromagnetic scattering characteristics of the turbulent wake were calculated using the two-scale facet mode. Using this method, the impact of different parameters on the scattering coefficient and the electromagnetic image of the turbulence wake were investigated, to explain the modulation mechanism and electromagnetic imaging characteristics of the near-field turbulent wake. Moreover, an application for estimating a ship’s heading is proposed based on the electromagnetic imaging characteristics of the turbulent wake. Full article
(This article belongs to the Topic High-Resolution Earth Observation Systems, Technologies, and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic of near-field turbulent wake on the (X, Y) plane.</p> Full article ">Figure 2
<p>Variations in parameters related to motion control: (<b>a</b>) magnitude of net force as a function of time; (<b>b</b>) relationship between propeller speeds <math display="inline"><semantics> <mi>n</mi> </semantics></math> and self-propelled speeds <math display="inline"><semantics> <mrow> <msub> <mi>U</mi> <mi>P</mi> </msub> </mrow> </semantics></math>.</p> Full article ">Figure 3
<p>Geometry of the KCS hull with KP505 propeller.</p> Full article ">Figure 4
<p>Grid division of KCS with KP505 propeller.</p> Full article ">Figure 5
<p>Verification based on half-width of the turbulent wake [<a href="#B21-remotesensing-13-05178" class="html-bibr">21</a>].</p> Full article ">Figure 6
<p>Turbulent wake at different values of <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> </mrow> </semantics></math>: (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>0.173</mn> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>0.097</mn> </mrow> </semantics></math>.</p> Full article ">Figure 7
<p>Wave height of turbulent wake at different values of <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> </mrow> </semantics></math>.</p> Full article ">Figure 8
<p>Comparison of wave heights of modulated and original sea surface.</p> Full article ">Figure 9
<p>Schematic of global and local coordinate systems.</p> Full article ">Figure 10
<p>Influence of different radar parameters on <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>σ</mi> </mrow> </semantics></math> in the backscattering observation, HH polarization: (<b>a</b>) incidence angle; (<b>b</b>) frequency.</p> Full article ">Figure 11
<p>Influence of wind speed and value of <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> </mrow> </semantics></math> on <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>σ</mi> </mrow> </semantics></math> in the backscattering observation, HH polarization: (<b>a</b>) wind speed; (<b>b</b>) value of <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> </mrow> </semantics></math>.</p> Full article ">Figure 12
<p>Backscattering coefficient distribution, HH polarization, wind speed is 3 <math display="inline"><semantics> <mrow> <msup> <mi>ms</mi> <mrow> <mo>−</mo> <mn>1</mn> </mrow> </msup> </mrow> </semantics></math>: (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>0.173</mn> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mi>J</mi> <mi>D</mi> </msub> <mo>=</mo> <mn>0.097</mn> </mrow> </semantics></math>.</p> Full article ">Figure 13
<p>Backscattering coefficient distribution, HH polarization, wind speed is 5 <math display="inline"><semantics> <mrow> <msup> <mi>ms</mi> <mrow> <mo>−</mo> <mn>1</mn> </mrow> </msup> </mrow> </semantics></math>: (<b>a</b>) 45°; (<b>b</b>) 90°; (<b>c</b>) 135°.</p> Full article ">Figure 14
<p>Backscattering coefficient distribution, HH polarization, wind speed is 5 <math display="inline"><semantics> <mrow> <msup> <mi>ms</mi> <mrow> <mo>−</mo> <mn>1</mn> </mrow> </msup> </mrow> </semantics></math>: (<b>a</b>) the imaging results; (<b>b</b>) the estimate of the ship heading based on the turbulent wake; (<b>c</b>) the estimate of the ship heading based on the Kelvin wake.</p> Full article ">
18 pages, 807 KiB  
Article
Mapping Large-Scale Forest Disturbance Types with Multi-Temporal CNN Framework
by Xi Chen, Wenzhi Zhao, Jiage Chen, Yang Qu, Dinghui Wu and Xuehong Chen
Remote Sens. 2021, 13(24), 5177; https://doi.org/10.3390/rs13245177 - 20 Dec 2021
Cited by 10 | Viewed by 3499
Abstract
Forests play a vital role in combating gradual developmental deficiencies and balancing regional ecosystems, yet they are constantly disturbed by man-made or natural events. Therefore, developing a timely and accurate forest disturbance detection strategy is urgently needed. The accuracy of traditional detection algorithms [...] Read more.
Forests play a vital role in combating gradual developmental deficiencies and balancing regional ecosystems, yet they are constantly disturbed by man-made or natural events. Therefore, developing a timely and accurate forest disturbance detection strategy is urgently needed. The accuracy of traditional detection algorithms depends on the selection of thresholds or the formulation of complete rules, which inevitably reduces the accuracy and automation level of detection. In this paper, we propose a new multitemporal convolutional network framework (MT-CNN). It is an integrated method that can realize long-term, large-scale forest interference detection and distinguish the types (forest fire and harvest/deforestation) of disturbances without human intervention. Firstly, it uses the sliding window technique to calculate an adaptive threshold to identify potential interference points, and then a multitemporal CNN network is designed to render the disturbance types with various disturbance duration periods. To illustrate the detection accuracy of MT-CNN, we conducted experiments in a large-scale forest area (about 990 km2) on the west coast of the United States (including northwest California and west Oregon) with long time-series Landsat data from 1986 to 2020. Based on the manually annotated labels, the evaluation results show that the overall accuracies of disturbance point detection and disturbance type recognition reach 90%. Also, this method is able to detect multiple disturbances that continuously occurred in the same pixel. Moreover, we found that forest disturbances that caused forest fire repeatedly appear without a significant coupling effect with annual temporal and precipitation variations. Potentially, our method is able to provide large-scale forest disturbance mapping with detailed disturbance information to support forest inventory management and sustainable development. Full article
(This article belongs to the Special Issue Remote Sensing and Geospatial Approaches for Studying the Environment Affected by Human Activities)
►▼ Show Figures

Figure 1

Figure 1
<p>Study on west coast area of USA. Area A is dominated by forest decline, while Area B is dominated by wildfires.</p> Full article ">Figure 2
<p>The typical time-series curves for different types of forest disturbances. (<b>a</b>) represents fire disturbance while (<b>b</b>) represents harvest/deforestation.</p> Full article ">Figure 3
<p>Two annotated areas with different types of forest disturbances. Harvest/deforestation disturbance samples are shown in <b>left</b>, while fire disturbance samples are shown in <b>right</b>.</p> Full article ">Figure 4
<p>Flowchart of proposed integrated MT-CNN. (<b>1</b>) yearly composite data are prepared for analysis; (<b>2</b>) disturbance point is detected with sliding window technique; (<b>3</b>) disturbance type recognition with multitemporal time-series fragments.</p> Full article ">Figure 5
<p>Flowchart of the multitemporal CNN framework.</p> Full article ">Figure 6
<p>(<b>a</b>–<b>c</b>) Disturbance mapping results on disturbance point, type, and frequency.</p> Full article ">Figure 7
<p>Finer-scale illustration of forest disturbance events with specific year.</p> Full article ">Figure 8
<p>Disturbance point detection results on area A and B. For area A, maps with forest fire in years of 1994–2002–2015 are illustrated. For area B, maps with harvest/deforestation in years of 1990–2000–2009 are illustrated.</p> Full article ">Figure 9
<p>Confusion matrix of disturbance point detection from 1986 to 2020.</p> Full article ">Figure 10
<p>Disturbance type identification on area A and area B. Left area B is dominated by forest fire, while right area A is dominated by harvest/deforestation.</p> Full article ">Figure 11
<p>Mapping results of forest disturbance frequencies. Left area B is dominated by 1 or 2 times disturbances while right area A dominated with more than 3 times number of forest disturbances within 35 years.</p> Full article ">Figure 12
<p>Statistical results on multiple disturbance with 5795 selected test samples.</p> Full article ">Figure 13
<p>Overall accuracy regarding disturbance types with different lengths of temporal fragment.</p> Full article ">Figure 14
<p>(<b>a</b>–<b>c</b>) Influences of terrain and weather factors on different types of forest disturbances.</p> Full article ">
20 pages, 5183 KiB  
Article
Monitoring Small Water Bodies Using High Spatial and Temporal Resolution Analysis Ready Datasets
by Vinicius Perin, Samapriya Roy, Joe Kington, Thomas Harris, Mirela G. Tulbure, Noah Stone, Torben Barsballe, Michele Reba and Mary A. Yaeger
Remote Sens. 2021, 13(24), 5176; https://doi.org/10.3390/rs13245176 - 20 Dec 2021
Cited by 9 | Viewed by 4104
Abstract
Basemap and Planet Fusion—derived from PlanetScope imagery—represent the next generation of analysis ready datasets that minimize the effects of the presence of clouds. These datasets have high spatial (3 m) and temporal (daily) resolution, which provides an unprecedented opportunity to improve the monitoring [...] Read more.
Basemap and Planet Fusion—derived from PlanetScope imagery—represent the next generation of analysis ready datasets that minimize the effects of the presence of clouds. These datasets have high spatial (3 m) and temporal (daily) resolution, which provides an unprecedented opportunity to improve the monitoring of on-farm reservoirs (OFRs)—small water bodies that store freshwater and play important role in surface hydrology and global irrigation activities. In this study, we assessed the usefulness of both datasets to monitor sub-weekly surface area changes of 340 OFRs in eastern Arkansas, USA, and we evaluated the datasets main differences when used to monitor OFRs. When comparing the OFRs surface area derived from Basemap and Planet Fusion to an independent validation dataset, both datasets had high agreement (r2 ≥ 0.87), and small uncertainties, with a mean absolute percent error (MAPE) between 7.05% and 10.08%. Pairwise surface area comparisons between the two datasets and the PlanetScope imagery showed that 61% of the OFRs had r2 ≥ 0.55, and 70% of the OFRs had MAPE <5%. In general, both datasets can be employed to monitor OFRs sub-weekly surface area changes, and Basemap had higher surface area variability and was more susceptible to the presence of cloud shadows and haze when compared to Planet Fusion, which had a smoother time series with less variability and fewer abrupt changes throughout the year. The uncertainties in surface area classification decreased as the OFRs increased in size. In addition, the surface area time series can have high variability, depending on the OFR environmental conditions (e.g., presence of vegetation inside the OFR). Our findings suggest that both datasets can be used to monitor OFRs sub-weekly, seasonal, and inter-annual surface area changes; therefore, these datasets can help improve freshwater management by allowing better assessment and management of the OFRs. Full article
(This article belongs to the Topic Water Management in the Era of Climatic Change)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study region in eastern Arkansas, USA, and the OFRs size distribution. The inset map represents the OFRs shapefile overlaid on SkySat satellite imagery.</p> Full article ">Figure 2
<p>Workflow used to estimate the OFRs’ surface area-time series from PlanetScope, Basemap, and Planet Fusion between July 2020 and July 2021.</p> Full article ">Figure 3
<p>Pairwise comparisons between the SkySat validation surface area and the surface area obtained from PlanetScope, Basemap, and Planet Fusion for multiple observations in time.</p> Full article ">Figure 4
<p>Number of surface area observations per month for PlanetScope and Basemap (<b>A</b>) and for Planet Fusion real, mixed, and synthetic (<b>C</b>). Frequency distribution of the total number of observations per OFR per year for PlanetScope and Basemap (<b>B</b>) and for Planet Fusion real, mixed, and synthetic (<b>D</b>).</p> Full article ">Figure 5
<p>Same-day pairwise comparisons between PlanetScope and Planet Fusion real, mixed, and synthetic for multiple observations in time and for all OFRs divided into three size classes (0.1–5 ha, 5–10 ha, and 10–30 ha). Brighter colors indicate higher point density.</p> Full article ">Figure 6
<p>Frequency distribution of r<sup>2</sup> and MAPE calculated by comparing the OFR time series from Basemap and Planet Fusion with PlanetScope.</p> Full article ">Figure 7
<p>Monthly percent error variability and MAPE calculated from the same-day pairwise comparisons between Basemap and Planet Fusion with PlanetScope for the three size classes (0.1–5 ha, 5–10 ha, and 10–30 ha).</p> Full article ">Figure 8
<p>OFRs (see <a href="#remotesensing-13-05176-t002" class="html-table">Table 2</a>) surface-area time series derived from PlanetScope, Basemap, and Planet Fusion and OFR shapefiles overlaid on high-resolution Google Satellite imagery.</p> Full article ">Figure 8 Cont.
<p>OFRs (see <a href="#remotesensing-13-05176-t002" class="html-table">Table 2</a>) surface-area time series derived from PlanetScope, Basemap, and Planet Fusion and OFR shapefiles overlaid on high-resolution Google Satellite imagery.</p> Full article ">Figure 9
<p>OFRs A and B (see <a href="#remotesensing-13-05176-t002" class="html-table">Table 2</a>) PlanetScope, Basemap, and Planet Fusion false-color composites (blue: red, green: green, and red: NIR) and the surface-water classification for 16 August 2020 (OFR A) and 30 August 2020 (OFR B).</p> Full article ">Figure 10
<p>OFR E (see <a href="#remotesensing-13-05176-t002" class="html-table">Table 2</a>) PlanetScope, Basemap, and Planet Fusion false-color composites (blue: red, green: green, and red: NIR) and the surface water classification for 14 July 2020 and 16 October 2020.</p> Full article ">
19 pages, 5547 KiB  
Article
Statistical Study of Ionospheric Equivalent Slab Thickness at Guam Magnetic Equatorial Location
by Yuqiang Zhang, Zhensen Wu, Jian Feng, Tong Xu, Zhongxin Deng, Ming Ou, Wen Xiong and Weimin Zhen
Remote Sens. 2021, 13(24), 5175; https://doi.org/10.3390/rs13245175 - 20 Dec 2021
Cited by 5 | Viewed by 2911
Abstract
The ionospheric equivalent slab thickness (τ) is defined as the ratio of the total electron content (TEC) to the F2-layer peak electron density (NmF2), and it is a significant parameter representative of the ionosphere. In this paper, a comprehensive statistical analysis [...] Read more.
The ionospheric equivalent slab thickness (τ) is defined as the ratio of the total electron content (TEC) to the F2-layer peak electron density (NmF2), and it is a significant parameter representative of the ionosphere. In this paper, a comprehensive statistical analysis of the diurnal, seasonal, solar, and magnetic activity variations in the τ at Guam (144.86°E, 13.62°N, 5.54°N dip lat), which is located near the magnetic equator, is presented using the GPS-TEC and ionosonde NmF2 data during the years 2012–2017. It is found that, for geomagnetically quiet days, the τ reaches its maximum value in the noontime, and the peak value in winter and at the equinox are larger than that in summer. Moreover, there is a post-sunset peak observed in the winter and equinox, and the τ during the post-midnight period is smallest in equinox. The mainly diurnal and seasonal variation of τ can be explained within the framework of relative variation of TEC and NmF2 during different seasonal local time. The dependence of τ on the solar activity shows positive correlation during the daytime, and the opposite situation applies for the nighttime. Specifically, the disturbance index (DI), which can visually assess the relationship between instantaneous τ values and the median, is introduced in the paper to quantitatively describe the overall pattern of the geomagnetic storm effect on the τ variation. The results show that the geomagnetic storm seems to have positive effect on the τ during most of the storm-time period at Guam. An example, on the 1 June 2013, is also presented to analyze the physical mechanism. During the positive storms, the penetration electric field, along with storm time equator-ward neutral wind, tends to increase upward drift and uplift F region, causing the large increase in TEC, accompanied by a relatively small increase in NmF2. On the other hand, an enhanced equatorward wind tends to push more plasma, at low latitudes, into the topside ionosphere in the equatorial region, resulting in the TEC not undergoing severe depletion, as with NmF2, during the negative storms. The results would complement the analysis of τ behavior during quiet and disturbed conditions at equatorial latitudes in East Asia. Full article
(This article belongs to the Special Issue Ionosphere Monitoring with Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>F10.7 variations during the years 2012–2017.</p> Full article ">Figure 2
<p>An example for the calculation of DI index during the period 6–8 June 2013.</p> Full article ">Figure 3
<p>Map of mean (<b>a</b>) and standard deviation (<b>b</b>) of equivalent slab thickness (τ) under geomagnetically quiet condition, according to January–December and 0LT–23LT division in 2012–2017.</p> Full article ">Figure 4
<p>Map of mean and standard deviation of equivalent slab thickness (τ) under geomagnetically quiet condition. (<b>a</b>) Mean of τ in high solar activity year 2014; (<b>b</b>) Mean of τ in low solar activity year 2016; (<b>c</b>) Standard deviation of τ in high solar activity year 2014; (<b>d</b>) Standard deviation of τ in low solar activity year 2016.</p> Full article ">Figure 4 Cont.
<p>Map of mean and standard deviation of equivalent slab thickness (τ) under geomagnetically quiet condition. (<b>a</b>) Mean of τ in high solar activity year 2014; (<b>b</b>) Mean of τ in low solar activity year 2016; (<b>c</b>) Standard deviation of τ in high solar activity year 2014; (<b>d</b>) Standard deviation of τ in low solar activity year 2016.</p> Full article ">Figure 5
<p>Diurnal variations of mean and standard deviation of equivalent slab thickness (τ) in winter, summer, and equinox at geomagnetically quiet condition.</p> Full article ">Figure 6
<p>Diurnal variation of the mean and standard deviation (STD) of DI index at geomagnetically quiet conditions.</p> Full article ">Figure 7
<p>Same as <a href="#remotesensing-13-05175-f006" class="html-fig">Figure 6</a> but for geomagnetic storm periods.</p> Full article ">Figure 8
<p>(<b>a</b>) Comparison of diurnal DI index for geomagnetically quiet condition and geomagnetic storm condition (<b>b</b>) The minus between the DI index during geomagnetic storm periods and DI index at geomagnetically quiet time, according to the sign of DI.</p> Full article ">Figure 9
<p>(<b>a</b>) Diurnal variations of mean and standard deviation of total electron content (TEC) in Guam during winter, summer, and equinox, respectively. (<b>b</b>) Diurnal variations of mean and standard deviation of F2-layer peak electron density (<span class="html-italic">Nm</span>F2) in Guam during winter, summer, and equinox, respectively.</p> Full article ">Figure 10
<p>(<b>a</b>) Variation rates in TEC for winter, summer, and equinox, respectively. (<b>b</b>) Variation rates in NmF2 for winter, summer, and equinox, respectively.</p> Full article ">Figure 11
<p>The geophysical conditions during 1–2 June 2013.</p> Full article ">Figure 12
<p>(<b>a</b>) The DI index of TEC and <span class="html-italic">Nm</span>F2 on 1 June 2013. (<b>b</b>) The DI index of τ on 1 June 2013.</p> Full article ">Figure 13
<p>The DI index of TTEC and BTEC on 1 June 2013.</p> Full article ">
24 pages, 84000 KiB  
Article
Real-Time Tephra Detection and Dispersal Forecasting by a Ground-Based Weather Radar
by Magfira Syarifuddin, Susanna F. Jenkins, Ratih Indri Hapsari, Qingyuan Yang, Benoit Taisne, Andika Bayu Aji, Nurnaning Aisyah, Hanggar Ganara Mawandha and Djoko Legono
Remote Sens. 2021, 13(24), 5174; https://doi.org/10.3390/rs13245174 - 20 Dec 2021
Cited by 1 | Viewed by 3206
Abstract
Tephra plumes can cause a significant hazard for surrounding towns, infrastructure, and air traffic. The current work presents the use of a small and compact X-band multi-parameter (X-MP) radar for the remote tephra detection and tracking of two eruptive events at Merapi Volcano, [...] Read more.
Tephra plumes can cause a significant hazard for surrounding towns, infrastructure, and air traffic. The current work presents the use of a small and compact X-band multi-parameter (X-MP) radar for the remote tephra detection and tracking of two eruptive events at Merapi Volcano, Indonesia, in May and June 2018. Tephra detection was performed by analysing the multiple parameters of radar: copolar correlation and reflectivity intensity factor. These parameters were used to cancel unwanted clutter and retrieve tephra properties, which are grain size and concentration. Real-time spatial and temporal forecasting of tephra dispersal was performed by applying an advection scheme (nowcasting) in the manner of an ensemble prediction system (EPS). Cross-validation was performed using field-survey data, radar observations, and Himawari-8 imageries. The nowcasting model computed both the displacement and growth and decaying rate of the plume based on the temporal changes in two-dimensional movement and tephra concentration, respectively. Our results are in agreement with ground-based data, where the radar-based estimated grain size distribution falls within the range of in situ grain size. The uncertainty of real-time forecasted tephra plume depends on the initial condition, which affects the growth and decaying rate estimation. The EPS improves the predictability rate by reducing the number of missed and false forecasted events. Our findings and the method presented here are suitable for early warning of tephra fall hazard at the local scale. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Left panel (<b>a</b>) is a fixed observation area of the PPI scan mode. Centre panel (<b>b</b>) is the 3D view of surface topography of the red square area in (<b>a</b>). Right panel (<b>c</b>) is the vertical profile of radar scan and topography, extracted along the blue arrow in (<b>a</b>). Identified in (<b>a</b>) are locations of Yogyakarta city, Merapi, and Merbabu. The locations of Turgo hill and Merapi summit are pointed by red arrows in (<b>b</b>) and the X-MP radar site is indicated by a red dot in (<b>a</b>,<b>b</b>).</p> Full article ">Figure 2
<p>Tephra detection using modified VARR model (<a href="#app2-remotesensing-13-05174" class="html-app">Appendix A </a>and <a href="#app3-remotesensing-13-05174" class="html-app"> Appendix B</a>) for M05 (first row) and M06 (bottom row) at 2 min after the reported eruption onset. From the left, the first column (<b>a</b>,<b>d</b>) is the reflectivity intensity factor, the second column (<b>b</b>,<b>e</b>) is the tephra classes, and the third column (<b>c</b>,<b>f</b>) is the tephra concentration. Tephra classes (<b>b</b>,<b>e</b>) of 1–3 represent the coarse class (C) of light, moderate, and intense concentration, respectively. Tephra classes of 4–6 represent the same tephra concentration, but for the finer class (F).</p> Full article ">Figure 3
<p>The first row is the spatial distribution of the estimated radar-retrieved mean diameter of M05 at different time stamps. The second row is the corresponding cumulative grain size (in ϕ) for each time stamp. In each image on the second row, the <span class="html-italic">x</span>-axis is the lower limit of ϕ = −log<sub>2</sub><span class="html-italic">D</span>.</p> Full article ">Figure 4
<p>The first row is the spatial distribution of the estimated radar-retrieved mean diameter of M06 at several time stamps. The second row is the corresponding cumulative grain size for each time stamp. In each image on the second row, the <span class="html-italic">x</span>-axis is the lower limit of ϕ = −log<sub>2</sub><span class="html-italic">D</span>.</p> Full article ">Figure 5
<p>Left image (<b>a</b>) is the location of in situ sampling points of grain size (ϕ) for both event cases, digitized from [<a href="#B24-remotesensing-13-05174" class="html-bibr">24</a>]. Available data in [<a href="#B24-remotesensing-13-05174" class="html-bibr">24</a>] are Q-01, Q-09, P-01, and P-09, as presented in the cumulative frequency of GSD in (<b>b</b>–<b>e</b>) as indicated in the top right panel. M05 is given by P-01 and P-09 and M06 is given by Q-01 and Q-09 (M06). The cumulative frequency GSDs in (<b>f</b>,<b>g</b>) are extracted from radar-retrieved GSD for Q-01 and Q-09 (M06), respectively. Panels (<b>h</b>,<b>i</b>) give the cumulative frequency of GSD M05 (data are merged from P-01, P-02, P-03, P-04, and P-05) and M06 (data are merged from Q-01, Q-02, Q-07, Q-08, and Q-09). The <span class="html-italic">x</span>-axis on the GSD histogram is the lower limit of ϕ = −log<sub>2</sub><span class="html-italic">D</span>.</p> Full article ">Figure 6
<p>The retrieved tephra concentration on M05 based on valid <math display="inline"><semantics> <mrow> <msub> <mi>Z</mi> <mi>H</mi> </msub> </mrow> </semantics></math> (<b>a</b>–<b>d</b>) compared with the mean EPS (<b>e</b>–<b>h</b>) at different time steps, as indicated in the top left corner. The mean EPS uses a threshold of tephra concentration ≥ 0.01 g/m<sup>3</sup>. In each image, the <span class="html-italic">x</span>-axis is longitude and <span class="html-italic">y</span>-axis is latitude.</p> Full article ">Figure 7
<p>The retrieved tephra concentration on M06 based on valid <math display="inline"><semantics> <mrow> <msub> <mi>Z</mi> <mi>H</mi> </msub> </mrow> </semantics></math> (<b>a</b>–<b>d</b>) compared with the mean EPS (<b>e</b>–<b>h</b>) at different time steps, as indicated in the top left corner. The mean EPS uses a threshold of tephra concentration ≥ 0.01 g/m<sup>3</sup>. In each image, the <span class="html-italic">x</span>-axis is longitude and <span class="html-italic">y</span>-axis is latitude.</p> Full article ">Figure 8
<p>Spaghetti plots of <span class="html-italic">CSI</span> (<b>left</b>) and <span class="html-italic">POD</span> (<b>right</b>) for both study cases. (<b>a</b>,<b>b</b>) M05 and (<b>c</b>,<b>d</b>) M06. The ensemble member was named based on IC: onset+<span class="html-italic">i</span>_Sc<span class="html-italic">j</span>, where <span class="html-italic">i</span> and <span class="html-italic">j</span> represent the time in minutes and advection phenomena (<a href="#remotesensing-13-05174-t002" class="html-table">Table 2</a>), respectively.</p> Full article ">Figure 9
<p>Brier score of tephra plume tracking by mean EPS of M05 and M06 based on the pair analysis with (<b>a</b>) radar and (<b>b</b>) Himawari-8. Each EPS was generated using six members.</p> Full article ">Figure 10
<p>Mean ensemble of tephra concentration and area of detected tephra cloud by Himawari-8 (given by black polygons). The study cases and time stamps are indicated in the top left corner. The detected area of Himawari-8′s tephra cloud was transformed into a polygon using cloud temperature &lt; 285 K. The ground-based data points are also presented in each map, given by black dots. In each image, the <span class="html-italic">x</span>- and <span class="html-italic">y</span>-axis are longitude and latitude, respectively.</p> Full article ">Figure A1
<p>An example of the clutter cancellation results. Data are presented for elevation angles of 15 and 18°. The cases and time stamps are given in the top left corner. (<b>a</b>,<b>d</b>) The observed radar reflectivity intensity factor <math display="inline"><semantics> <mrow> <msub> <mi>Z</mi> <mi>H</mi> </msub> </mrow> </semantics></math><span class="html-italic"><sub>H</sub></span>, (<b>b</b>,<b>e</b>) the filtered <math display="inline"><semantics> <mrow> <msub> <mi>Z</mi> <mi>H</mi> </msub> </mrow> </semantics></math> after applying the clutter cancellation procedure, and (<b>c</b>,<b>f</b>) the frequency clutter map <span class="html-italic">Pc</span> for the given elevation angles. In each image, the <span class="html-italic">x</span>-axis is longitude and <span class="html-italic">y</span>-axis is latitude.</p> Full article ">Figure A2
<p>The bright temperature difference of band-13 (BTD<sub>13</sub>) of Himawari-8. Data are presented here for M05 (<b>left</b>) and M06 (<b>right</b>), 30 min after reported eruption onset, as indicated in the top left corner.</p> Full article ">Figure A3
<p>Some examples of the ensemble members (row) at different time step output (column). Presented is the nowcasting of M05. Each ensemble member was named based on the initial condition: onset+<span class="html-italic">i</span>_Sc<span class="html-italic">j</span>, where <span class="html-italic">i</span> and <span class="html-italic">j</span> represent the time in minutes and advection scenario (<a href="#remotesensing-13-05174-t003" class="html-table">Table 3</a>), respectively. In each image, the <span class="html-italic">x</span>-axis is longitude and <span class="html-italic">y</span>-axis is latitude.</p> Full article ">Figure A4
<p>Some examples of the ensemble members (row) at different time step output (column). Presented is the nowcasting of M06. Each ensemble member was named based on the initial condition: onset+<span class="html-italic">i</span>_Sc<span class="html-italic">j</span>, where <span class="html-italic">i</span> and <span class="html-italic">j</span> represent the time in minutes and advection scenario (<a href="#remotesensing-13-05174-t003" class="html-table">Table 3</a>), respectively. In each image, the <span class="html-italic">x</span>-axis is longitude and <span class="html-italic">y</span>-axis is latitude.</p> Full article ">Figure A5
<p>Number of missed and false events of ensemble members and the mean ensemble for M05 (<b>a</b>) and M06 (<b>b</b>).</p> Full article ">
21 pages, 9352 KiB  
Article
A Comparison of UAV RGB and Multispectral Imaging in Phenotyping for Stay Green of Wheat Population
by Xiaofeng Cao, Yulin Liu, Rui Yu, Dejun Han and Baofeng Su
Remote Sens. 2021, 13(24), 5173; https://doi.org/10.3390/rs13245173 - 20 Dec 2021
Cited by 18 | Viewed by 4441
Abstract
High throughput phenotyping (HTP) for wheat (Triticum aestivum L.) stay green (SG) is expected in field breeding as SG is a beneficial phenotype for wheat high yield and environment adaptability. The RGB and multispectral imaging based on the unmanned aerial vehicle (UAV) [...] Read more.
High throughput phenotyping (HTP) for wheat (Triticum aestivum L.) stay green (SG) is expected in field breeding as SG is a beneficial phenotype for wheat high yield and environment adaptability. The RGB and multispectral imaging based on the unmanned aerial vehicle (UAV) are widely popular multi-purpose HTP platforms for crops in the field. The purpose of this study was to compare the potential of UAV RGB and multispectral images (MSI) in SG phenotyping of diversified wheat germplasm. The multi-temporal images of 450 samples (406 wheat genotypes) were obtained and the color indices (CIs) from RGB and MSI and spectral indices (SIs) from MSI were extracted, respectively. The four indices (CIs in RGB, CIs in MSI, SIs in MSI, and CIs + SIs in MSI) were used to detect four SG stages, respectively, by machine learning classifiers. Then, all indices’ dynamics were analyzed and the indices that varied monotonously and significantly were chosen to calculate wheat temporal stay green rates (SGR) to quantify the SG in diverse genotypes. The correlations between indices’ SGR and wheat yield were assessed and the dynamics of some indices’ SGR with different yield correlations were tracked in three visual observed SG grades samples. In SG stage detection, classifiers best average accuracy reached 93.20–98.60% and 93.80–98.80% in train and test set, respectively, and the SIs containing red edge or near-infrared band were more effective than the CIs calculated only by visible bands. Indices’ temporal SGR could quantify SG changes on a population level, but showed some differences in the correlation with yield and in tracking visual SG grades samples. In SIs, the SGR of Normalized Difference Red-edge Index (NDRE), Red-edge Chlorophyll Index (CIRE), and Normalized Difference Vegetation Index (NDVI) in MSI showed high correlations with yield and could track visual SG grades at an earlier stage of grain filling. In CIs, the SGR of Normalized Green Red Difference Index (NGRDI), the Green Leaf Index (GLI) in RGB and MSI showed low correlations with yield and could only track visual SG grades at late grain filling stage and that of Norm Red (NormR) in RGB images failed to track visual SG grades. This study preliminarily confirms the MSI is more available and reliable than RGB in phenotyping for wheat SG. The index-based SGR in this study could act as HTP reference solutions for SG in diversified wheat genotypes. Full article
(This article belongs to the Section Remote Sensing in Agriculture and Vegetation)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental site; (<b>A</b>) Shaanxi Province, China; (<b>B</b>) Cao Xinzhuang Farm in Yangling Zone.</p> Full article ">Figure 2
<p>Daily air temperature during the test period.</p> Full article ">Figure 3
<p>Classifier confusion matrices; (<b>A</b>) SVM built by CIs or SIs; (<b>B</b>) QDA built by CIs or SIs; (<b>C</b>) KNN built by CIs or SIs; (<b>D</b>) EL built by CIs or SIs.</p> Full article ">Figure 4
<p>(<b>A</b>–<b>D</b>) Time series shadow error curves of indices in RGB or MSI.</p> Full article ">Figure 5
<p>Index correlation coefficient matrices; (<b>A</b>) RGB_CIs; (<b>B</b>) MSI_CIs; (<b>C</b>) MSI_SIs.</p> Full article ">Figure 6
<p>Calibration and prediction set samples yield distribution.</p> Full article ">Figure 7
<p>Scatter plots of measured and predicted yield in MLR models of some indices’ SGR. (<b>A</b>) SG_NDRE-MLR; (<b>B</b>) SG_CIRE-MLR; (<b>C</b>) SG_NormR_MSI-MLR; (<b>D</b>) SG_NormR_RGB-MLR; (<b>E</b>) SG_NGRDI_MSI-MLR; (<b>F</b>) SG_NGRDI_RGB-MLR.</p> Full article ">Figure 8
<p>(<b>A</b>–<b>J</b>) The dynamics of some indices’ SGRs in RGB or MSI in three grade samples.</p> Full article ">Figure 9
<p>Digital images of selected samples at four key stages.</p> Full article ">Figure 10
<p>The dynamics of some indices’ SGRs in RGB or MSI in selected samples.</p> Full article ">
16 pages, 3704 KiB  
Article
A Four-Step Method for Estimating Suspended Particle Size Based on In Situ Comprehensive Observations in the Pearl River Estuary in China
by Zuomin Wang, Shuibo Hu, Qingquan Li, Huizeng Liu, Xiaomei Liao and Guofeng Wu
Remote Sens. 2021, 13(24), 5172; https://doi.org/10.3390/rs13245172 - 20 Dec 2021
Cited by 6 | Viewed by 3115
Abstract
The suspended particle size has great impacts on marine biology environments and biogeochemical processes, such as the settling rates of particles and sunlight transmission in marine water. However, the spatial–temporal variations in particle sizes in coastal waters are rarely reported due to the [...] Read more.
The suspended particle size has great impacts on marine biology environments and biogeochemical processes, such as the settling rates of particles and sunlight transmission in marine water. However, the spatial–temporal variations in particle sizes in coastal waters are rarely reported due to the paucity of appropriate observations and the limitations of particle size retrieval methods, especially in areas with complex optical properties. This study proposed a remote sensing-based method for estimating the median particle size Dv50 (calculated with a size range of 2.05–297 μm) that correlates Dv50 with the inherent optical properties (IOPs) retrieved from in situ remote sensing reflectance above the water’s surface (Rrs(λ)) in the Pearl River estuary (PRE) in China. Rrs(λ) was resampled to simulate the Multispectral Instrument (MSI) onboard Sentinel-2A/B, and the wavebands in 490, 560, and 705 nm were utilized for the retrieval of the IOPs. The results of this method had a statistical performance of 0.86, 18.52, 21.28%, and −1.85 for the R2, RMSE, MAPE, and bias values, respectively, in validation, which indicated that Dv50 could be estimated by Rrs(λ) with the proposed four-step method. Then, the proposed method was applied to Sentinel-2 MSI imagery, and a clear difference in Dv50 distribution which was retrieved from a different time could be seen. The proposed method holds great potential for monitoring the suspended particle size of coastal waters. Full article
(This article belongs to the Special Issue Remote Sensing Monitoring of Ocean and Coastal Biogeochemistry)
►▼ Show Figures

Figure 1

Figure 1
<p>Study area and sampling sites.</p> Full article ">Figure 2
<p>D<sub>v</sub><sup>50</sup> retrieval schematic flow chart.</p> Full article ">Figure 3
<p>(<b>a</b>) Validation of QAA method for b<sub>bp</sub>(532) retrieval. (<b>b</b>) The relationship between b<sub>bp</sub>(532) and b<sub>p</sub>(532).</p> Full article ">Figure 4
<p>The V<sub>tot</sub>(D) estimation from (<b>a</b>) b<sub>p</sub>(532) and (<b>b</b>) b<sub>bp</sub>(532).</p> Full article ">Figure 5
<p>The performance of D<sub>v</sub><sup>50</sup> retrieval models with b<sub>p</sub>*(532) ((<b>a</b>) inverse proportion, (<b>c</b>) negative power function, and (<b>e</b>) negative exponential function) and b<sub>bp</sub><sup>*</sup>(532) ((<b>b</b>) inverse proportion, (<b>d</b>) negative power function, and (<b>f</b>) negative exponential function).</p> Full article ">Figure 6
<p>Validation of the D<sub>v</sub><sup>50</sup> retrieval models proposed by previous studies and developed in this study.</p> Full article ">Figure 7
<p>The D<sub>v</sub><sup>50</sup> distribution retrieved by the proposed method on (<b>a</b>) 22 September 2019 and (<b>b</b>) 30 January 2020.</p> Full article ">Figure 8
<p>(<b>a</b>) Correlations between V(D) and b<sub>p</sub>(532), b<sub>bp</sub>(532), and c<sub>p</sub>(670) in all size ranges (the gray region shows the particle size range within the correlation coefficient r &gt; 0.8 for b<sub>p</sub>(532)). (<b>b</b>) The “Junge distribution” for the particle number concentration (the blue line represents the fitted line of the median PSD slope (ξ = 3.55) of all the data).</p> Full article ">
19 pages, 19674 KiB  
Article
Protection Effect and Vacancy of the Ecological Protection Redline: A Case Study in Guangdong–Hong Kong–Macao Greater Bay Area, China
by Xiuming Wang, Youyue Wen, Xucheng Liu, Ding Wen, Yingxian Long, Peng Zhao, Piao Liu and Jenny Zhong
Remote Sens. 2021, 13(24), 5171; https://doi.org/10.3390/rs13245171 - 20 Dec 2021
Cited by 20 | Viewed by 4130
Abstract
The Ecological Protection Redline (EPR) is an innovative measure implemented in China to maintain the structural stability and functional security of the ecosystem. By prohibiting large-scale urban and industrial construction activities, EPR is regarded as the “lifeline” to ensure national ecological security. It [...] Read more.
The Ecological Protection Redline (EPR) is an innovative measure implemented in China to maintain the structural stability and functional security of the ecosystem. By prohibiting large-scale urban and industrial construction activities, EPR is regarded as the “lifeline” to ensure national ecological security. It is of great practical significance to scientifically evaluate the protection effect of EPR and identify the protection vacancies. However, current research has focused only on the protection effects of the EPR on ecosystem services (ESs), and the protection effect of the EPR on ecological connectivity remains poorly understood. Based on an evaluation of ES importance, the circuit model, and hotspot analysis, this paper identified the ecological security pattern in Guangdong–Hong Kong–Macao Greater Bay Area (GBA), analyzed the role of EPR in maintaining ES and ecological connectivity, and identified protection gaps. The results were as follows: (1) The ecological sources were mainly distributed in mountainous areas of the GBA. The ecological sources and ecological corridors constitute a circular ecological shelter surrounding the urban agglomeration of the GBA. (2) The EPR effectively protected water conservation, soil conservation, and biodiversity maintenance services, but the protection efficiency of carbon sequestration service and ecological connectivity were low. In particularly, EPR failed to continuously protect regional large-scale ecological corridors and some important stepping stones. (3) The protection gaps of carbon sequestration service and ecological connectivity in the study area reached 1099.80 km2 and 2175.77 km2, respectively, mainly distributed in Qingyuan, Yunfu, and Huizhou. In future EPR adjustments, important areas for carbon sequestration service and ecological connectivity maintenance should be included. This study provides a comprehensive understanding of the protection effects of EPR on ecological structure and function, and it has produced significant insights into improvements of the EPR policy. In addition, this paper proposes that the scope of resistance surface should be extended, which would improve the rationality of the ecological corridor simulation. Full article
(This article belongs to the Special Issue Remote Sensing of Ecosystem Functions: Advances for Ecosystems Conservation Status Assessment)
►▼ Show Figures

Figure 1

Figure 1
<p>The study area. (<b>a</b>) Location of GBA in China. (<b>b</b>) The EPR in the study area.</p> Full article ">Figure 2
<p>The technology roadmap.</p> Full article ">Figure 3
<p>ES importance and ecological sources. (<b>a</b>) Water conservation service. (<b>b</b>) Soil conservation service. (<b>c</b>) Biodiversity maintenance service. (<b>d</b>) Carbon sequestration service. (<b>e</b>) Distribution of ecological sources.</p> Full article ">Figure 3 Cont.
<p>ES importance and ecological sources. (<b>a</b>) Water conservation service. (<b>b</b>) Soil conservation service. (<b>c</b>) Biodiversity maintenance service. (<b>d</b>) Carbon sequestration service. (<b>e</b>) Distribution of ecological sources.</p> Full article ">Figure 4
<p>Ecological flow based on different ranges of resistance surfaces. (<b>a</b>) Based on a 10 km buffer out of the GBA. (<b>b</b>) Based on a 50 km buffer out of the GBA. (<b>c</b>) Based on a 200 km buffer out of the GBA.</p> Full article ">Figure 4 Cont.
<p>Ecological flow based on different ranges of resistance surfaces. (<b>a</b>) Based on a 10 km buffer out of the GBA. (<b>b</b>) Based on a 50 km buffer out of the GBA. (<b>c</b>) Based on a 200 km buffer out of the GBA.</p> Full article ">Figure 5
<p>Ecological corridors and ecological security patterns. (<b>a</b>) Simulation results of the current. (<b>b</b>) Ecological security patterns.</p> Full article ">Figure 5 Cont.
<p>Ecological corridors and ecological security patterns. (<b>a</b>) Simulation results of the current. (<b>b</b>) Ecological security patterns.</p> Full article ">Figure 6
<p>Spatial relationship between EPR and ecological corridors within 50 km buffer zone of the GBA. (<b>a</b>) EPR and current simulation results. (<b>b</b>) EPR and total current of each corridor. (<b>c</b>) EPR and average current of each corridor. (<b>d</b>) Stepping stones not protected by EPR.</p> Full article ">Figure 6 Cont.
<p>Spatial relationship between EPR and ecological corridors within 50 km buffer zone of the GBA. (<b>a</b>) EPR and current simulation results. (<b>b</b>) EPR and total current of each corridor. (<b>c</b>) EPR and average current of each corridor. (<b>d</b>) Stepping stones not protected by EPR.</p> Full article ">Figure 7
<p>Protection vacancies: (<b>a</b>) protection vacancies of carbon sequestration, (<b>b</b>) protection vacancies that can improve the efficiency of connectivity protection, (<b>c</b>) protection vacancies that played an important role in maintaining corridor patency.</p> Full article ">Figure 7 Cont.
<p>Protection vacancies: (<b>a</b>) protection vacancies of carbon sequestration, (<b>b</b>) protection vacancies that can improve the efficiency of connectivity protection, (<b>c</b>) protection vacancies that played an important role in maintaining corridor patency.</p> Full article ">
26 pages, 2191 KiB  
Article
Estimating Stand and Fire-Related Surface and Canopy Fuel Variables in Pine Stands Using Low-Density Airborne and Single-Scan Terrestrial Laser Scanning Data
by Cecilia Alonso-Rego, Stéfano Arellano-Pérez, Juan Guerra-Hernández, Juan Alberto Molina-Valero, Adela Martínez-Calvo, César Pérez-Cruzado, Fernando Castedo-Dorado, Eduardo González-Ferreiro, Juan Gabriel Álvarez-González and Ana Daría Ruiz-González
Remote Sens. 2021, 13(24), 5170; https://doi.org/10.3390/rs13245170 - 20 Dec 2021
Cited by 21 | Viewed by 3697
Abstract
In this study, we used data from a thinning trial conducted on 34 different sites and 102 sample plots established in pure and even-aged Pinus radiata and Pinus pinaster stands, to test the potential use of low-density airborne laser scanning (ALS) metrics and [...] Read more.
In this study, we used data from a thinning trial conducted on 34 different sites and 102 sample plots established in pure and even-aged Pinus radiata and Pinus pinaster stands, to test the potential use of low-density airborne laser scanning (ALS) metrics and terrestrial laser scanning (TLS) metrics to provide accurate estimates of variables related to surface and canopy fires. An exhaustive field inventory was carried out in each plot to estimate the main stand variables and the main variables related to fire hazard: surface fuel loads by layers, fuel strata gap, surface fuel height, stand mean height, canopy base height, canopy fuel load and canopy bulk density. In addition, the point clouds from low-density ALS and single-scan TLS of each sample plot were used to calculate metrics related to the vertical and horizontal distribution of forest fuels. The comparative performance of the following three non-parametric machine learning techniques used to estimate the main stand- and fire-related variables from those metrics was evaluated: (i) multivariate adaptive regression splines (MARS), (ii) support vector machine (SVM), and (iii) random forest (RF). The selection of the best modeling approach was based on a comparison of the root mean square error (RMSE), obtained by optimizing the parameters of each technique and performing cross-validation. Overall, the best results were obtained with the MARS techniques for data from both sensors. The TLS data provided the best results for variables associated with the internal characteristics of canopy structure and understory fuel but were less reliable for estimating variables associated with the upper canopy, due to occlusion by mid-canopy foliage. The combination of ALS and TLS metrics improved the accuracy of estimates for all variables analyzed, except the height and the biomass of the understory shrubs. The variability demonstrated by the combined use of both types of metrics ranged from 43.11% for the biomass of duff litter layers to 94.25% for dominant height. The results suggest that the combination of machine learning techniques and metrics derived from low-density ALS data, drawn from a single-scan TLS or a combination of both metrics, may represent a promising alternative to traditional field inventories for obtaining valuable information about surface and canopy fuel variables at large scales. Full article
(This article belongs to the Special Issue 3D Point Clouds in Forest Remote Sensing II)
►▼ Show Figures

Figure 1

Figure 1
<p>Geographical location of the 34 thinning trials in the study area (<b>left</b>). Example of three sample plots (green polygons) in one of the trial locations (<b>middle right</b>). Arrangement of surface fuel transects in each sample plot (<b>lower right</b>).</p> Full article ">Figure 2
<p>Relative importance of independent variables for the best-fit models for estimating surface fire-related variables (<b>upper</b>), canopy fire-related variables (<b>middle</b>) and stand variables (<b>lower</b>), respectively. The values were estimated from the RMSE reduction that the incorporation of each variable in the model implies, assuming that the other independent variables were already included in the model.</p> Full article ">Figure 2 Cont.
<p>Relative importance of independent variables for the best-fit models for estimating surface fire-related variables (<b>upper</b>), canopy fire-related variables (<b>middle</b>) and stand variables (<b>lower</b>), respectively. The values were estimated from the RMSE reduction that the incorporation of each variable in the model implies, assuming that the other independent variables were already included in the model.</p> Full article ">Figure A1
<p>Plots of observed versus predicted values with the best-fitted models for estimating surface and canopy fire-related and stand variables. The red line represents the linear relationship between the two types of values.</p> Full article ">Figure A1 Cont.
<p>Plots of observed versus predicted values with the best-fitted models for estimating surface and canopy fire-related and stand variables. The red line represents the linear relationship between the two types of values.</p> Full article ">Figure A1 Cont.
<p>Plots of observed versus predicted values with the best-fitted models for estimating surface and canopy fire-related and stand variables. The red line represents the linear relationship between the two types of values.</p> Full article ">
11 pages, 2946 KiB  
Technical Note
Statistical Analysis for Tidal Flat Classification and Topography Using Multitemporal SAR Backscattering Coefficients
by Keunyong Kim, Hahn Chul Jung, Jong-Kuk Choi and Joo-Hyung Ryu
Remote Sens. 2021, 13(24), 5169; https://doi.org/10.3390/rs13245169 - 20 Dec 2021
Cited by 5 | Viewed by 3688
Abstract
Coastal zones are very dynamic natural systems that experience short-term and long-term morphological changes. Their highly dynamic behavior requires frequent monitoring. Tidal flat topography for a large spatial coverage has been generated mainly by the waterline extraction method from multitemporal remote sensing observations. [...] Read more.
Coastal zones are very dynamic natural systems that experience short-term and long-term morphological changes. Their highly dynamic behavior requires frequent monitoring. Tidal flat topography for a large spatial coverage has been generated mainly by the waterline extraction method from multitemporal remote sensing observations. Despite the efficiency and robustness of the waterline extraction method, the waterline-based digital elevation model (DEM) is limited to representing small scale topographic features, such as localized tidal tributaries. Tidal flats show a rapid increase in SAR backscattering coefficients when the tide height is lower than the tidal flat topography compared to when the tidal flat is covered by water. This leads to a tidal flat with a distinct statistical behavior on the temporal variability of our multitemporal SAR backscattering coefficients. Therefore, this study aims to suggest a new method that can overcome the constraints of the waterline-based method by using a pixel-based DEM generation algorithm. Jenks Natural Break (JNB) optimization was applied to distinguish the tidal flat from land and ocean using multitemporal Senitnel-1 SAR data for the years 2014–2020. We also implemented a logistic model to characterize the temporal evolution of the SAR backscattering coefficients along with the tide heights and estimated intertidal topography. The Sentinel-1 DEM from the JNB classification and logistic function was evaluated by an airborne Lidar DEM. Our pixel-based DEM outperformed the waterline-based Landsat DEM. This study demonstrates that our statistical approach to intertidal classification and topography serves to monitor the near real-time spatiotemporal distribution changes of tidal flats through continuous and stable SAR data collection on local and regional scales. Full article
(This article belongs to the Special Issue Advances in Spaceborne SAR – Technology and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Study area of the Hwangdo tidal flat in Cheonsu Bay overlaid with the SRTM elevation maps. The gray box shows the Sentinel-1 SAR coverage in path 127. The red circle shows a tide gauge. (<b>b</b>) Google Earth image of the Hwangdo tidal flat.</p> Full article ">Figure 2
<p>Temporal distribution of tide heights from October 2014 to July 2020. The gray line shows tide heights. The blue dots denote the acquisition times of the Sentinel-1 SAR images used in our tidal flat topography (MHHW: mean higher high water, MLLW: mean lower low water, HAT: highest astronomical tides, and LAT: lowest astronomical tides).</p> Full article ">Figure 3
<p>Spatial distribution of (<b>a</b>) the mean, (<b>b</b>) standard deviations, and (<b>c</b>) GVF of the Sentinel-1 SAR backscattering coefficients. (<b>d</b>) Estimated intertidal topography in areas of GVF values higher than 0.2. The labelled symbols denote examples of ocean, salt pond, land, and tidal flats used for the best fitting logistic model in <a href="#remotesensing-13-05169-f004" class="html-fig">Figure 4</a>.</p> Full article ">Figure 4
<p>Temporal variability of the radar backscattering coefficient gamma naught value versus tide height (<b>a</b>) ocean, (<b>b</b>) salt pond, (<b>c</b>) land, and (<b>d</b>–<b>f</b>) tidal flats with the best fitting logistic function as red solid lines.</p> Full article ">Figure 5
<p>The RMSE and MAE of the Sentinel-1 DEM against the reference Lidar DEM for each 0.1 bins of incremental GVF value.</p> Full article ">Figure 6
<p>The comparison of different DEM products from (<b>a</b>) the airborne Lidar, (<b>b</b>) the pixel-based Sentinel-1, and (<b>c</b>) the waterline-based Landsat ETM+ data. The solid white line indicates the topographic profiles in <a href="#remotesensing-13-05169-f007" class="html-fig">Figure 7</a>.</p> Full article ">Figure 7
<p>The topographic profiles of the airborne Lidar, pixel-based Sentinel-1, and waterline-based Landsat DEMs along the solid white line in <a href="#remotesensing-13-05169-f006" class="html-fig">Figure 6</a>.</p> Full article ">
14 pages, 5840 KiB  
Article
Investigation of Vegetation Changes in Different Mining Areas in Liaoning Province, China, Using Multisource Remote Sensing Data
by Baodong Ma, Xiangru Yang, Yajiao Yu, Yang Shu and Defu Che
Remote Sens. 2021, 13(24), 5168; https://doi.org/10.3390/rs13245168 - 20 Dec 2021
Cited by 12 | Viewed by 3660
Abstract
Mining can provide necessary mineral resources for humans. However, mining activities may cause damage to the surrounding ecology and environment. Vegetation change analysis is a key tool for evaluating damage to ecology and the environment. Liaoning is one of the major mining provinces [...] Read more.
Mining can provide necessary mineral resources for humans. However, mining activities may cause damage to the surrounding ecology and environment. Vegetation change analysis is a key tool for evaluating damage to ecology and the environment. Liaoning is one of the major mining provinces in China, with rich mineral resources and long-term, high-intensity mining activities. Taking Liaoning Province as an example, vegetation change in six mining areas was investigated using multisource remote sensing data to evaluate ecological and environmental changes. Based on MODIS NDVI series data from 2000 to 2019, change trends of vegetation were evaluated using linear regression. According to the results, there are large highly degraded vegetation areas in the Anshan, Benxi, and Yingkou mining areas, which indicates that mining activities have seriously damaged the vegetation in these areas. In contrast, there are considerable areas with improved vegetation in the Anshan, Fushun, and Fuxin mining areas, which indicates that ecological reclamation has played a positive role in these areas. Based on Sentinel-2A data, leaf chlorophyll content was inferred by using the vegetation index MERIS Terrestrial Chlorophyll Index (MTCI) after measurement of leaf spectra and chlorophyll content were carried out on the ground to validate the performance of MTCI. According to the results, the leaf chlorophyll content in the mines is generally lower than in adjacent areas in these mining areas with individual differences. In the Yingkou mining area, the chlorophyll content in adjacent areas is close to the magnesite mines, which means the spillover effect of environmental pollution in mines should be considerable. In the Anshan, Benxi, and Diaobingshan mining areas, the environmental stress on adjacent areas is slight. All in all, iron and magnesite open-pit mines should be monitored closely for vegetation destruction and stress due to the high intensity of mining activities and serious pollution. In contrast, the disturbance to vegetation is limited in resource-exhausted open-pit coal mines and underground coal mines. It is suggested that land reclamation should be enhanced to improve the vegetation in active open-pit mining areas, such as the Anshan, Benxi, and Yingkou mining areas. Additionally, environmental protection measures should be enhanced to relieve vegetation stress in the Yingkou mining area. Full article
(This article belongs to the Special Issue The Impact of Extreme Climatic and Disturbance Events on Vegetation Using Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>The location of six mining areas in Liaoning Province, China: (<b>A</b>) is Anshan mining area, (<b>B</b>) is Benxi mining area, (<b>C</b>) is Diaobingshan mining area, (<b>D</b>) is Fushun mining area, (<b>E</b>) is Fuxin mining area, and (<b>F</b>) is Yingkou mining area. The images are RGB with band 11, 8 and 4 of Sentinel-2A images.</p> Full article ">Figure 2
<p>The trend of vegetation indices over in the entire Liaoning Province and six mining areas from 2000 to 2019.</p> Full article ">Figure 3
<p>The spatial distribution of pixel-level change trend of vegetation from 2000 to 2019: (<b>A</b>) is Anshan mining area, (<b>B</b>) is Benxi mining area, (<b>C</b>) is Diaobingshan mining area, (<b>D</b>) is Fushun mining area, (<b>E</b>) is Fuxin mining area, and (<b>F</b>) is Yingkou mining area.</p> Full article ">Figure 4
<p>Relationship between VIs and leaf chlorophyll content: (<b>a</b>) is MTCI, and (<b>b</b>) is NDVI.</p> Full article ">Figure 5
<p>Spatial distribution of leaf chlorophyll content in the mining areas: (<b>A</b>) is Anshan mining area, (<b>B</b>) is Diaobingshan mining area, (<b>C</b>) is Benxi mining area, (<b>D</b>) is Fushun mining area, (<b>E</b>) is Fuxin mining area, and (<b>F</b>) is Yingkou mining area.</p> Full article ">Figure 6
<p>Histogram of chlorophyll content pixel for mine and adjacent areas in the mining area.</p> Full article ">Figure 7
<p>Chlorophyll content ratio of mines to adjacent areas in each mining area.</p> Full article ">Figure 8
<p>Vegetation improvement in open-pit mines based on MODIS NDVI and GeoEye image: (<b>a</b>) is in Anshan area, (<b>b</b>) is in Fushun area, and (<b>c</b>) is in Fuxin area.</p> Full article ">Figure 9
<p>Environmental pollution in mining areas: (<b>a</b>) is water pollution, and (<b>b</b>) is dust pollution.</p> Full article ">
27 pages, 10840 KiB  
Article
Estimating Actual Evapotranspiration over Croplands Using Vegetation Index Methods and Dynamic Harvested Area
by Neda Abbasi, Hamideh Nouri, Kamel Didan, Armando Barreto-Muñoz, Sattar Chavoshi Borujeni, Hamidreza Salemi, Christian Opp, Stefan Siebert and Pamela Nagler
Remote Sens. 2021, 13(24), 5167; https://doi.org/10.3390/rs13245167 - 20 Dec 2021
Cited by 23 | Viewed by 5086
Abstract
Advances in estimating actual evapotranspiration (ETa) with remote sensing (RS) have contributed to improving hydrological, agricultural, and climatological studies. In this study, we evaluated the applicability of Vegetation-Index (VI) -based ETa (ET-VI) for mapping and monitoring drought in arid agricultural systems in a [...] Read more.
Advances in estimating actual evapotranspiration (ETa) with remote sensing (RS) have contributed to improving hydrological, agricultural, and climatological studies. In this study, we evaluated the applicability of Vegetation-Index (VI) -based ETa (ET-VI) for mapping and monitoring drought in arid agricultural systems in a region where a lack of ground data hampers ETa work. To map ETa (2000–2019), ET-VIs were translated and localized using Landsat-derived 3- and 2-band Enhanced Vegetation Indices (EVI and EVI2) over croplands in the Zayandehrud River Basin (ZRB) in Iran. Since EVI and EVI2 were optimized for the MODerate Imaging Spectroradiometer (MODIS), using these VIs with Landsat sensors required a cross-sensor transformation to allow for their use in the ET-VI algorithm. The before- and after- impact of applying these empirical translation methods on the ETa estimations was examined. We also compared the effect of cropping patterns’ interannual change on the annual ETa rate using the maximum Normalized Difference Vegetation Index (NDVI) time series. The performance of the different ET-VIs products was then evaluated. Our results show that ETa estimates agreed well with each other and are all suitable to monitor ETa in the ZRB. Compared to ETc values, ETa estimations from MODIS-based continuity corrected Landsat-EVI (EVI2) (EVIMccL and EVI2MccL) performed slightly better across croplands than those of Landsat-EVI (EVI2) without transformation. The analysis of harvested areas and ET-VIs anomalies revealed a decline in the extent of cultivated areas and a loss of corresponding water resources downstream. The findings show the importance of continuity correction across sensors when using empirical algorithms designed and optimized for specific sensors. Our comprehensive ETa estimation of agricultural water use at 30 m spatial resolution provides an inexpensive monitoring tool for cropping areas and their water consumption. Full article
(This article belongs to the Special Issue Agriculture: Management, Disturbance, and Climate around the World Using the Google Earth Engine)
►▼ Show Figures

Figure 1

Figure 1
<p>Study area, (<b>a</b>) an overview of the study area: vegetated areas, the aquifer, irrigation networks, and the long-term mean annual precipitation sum (2003 to 2018). (<b>b</b>) Köppen–Geiger climate zones using the same color, BWh: Arid, desert, hot; BWk: Arid, desert, cold; BSk: Arid, steppe, cold; Dsa: Cold, dry summer, hot summer; Dsb: Cold, dry summer, warm summer; counties: 1: Tiran, 2: Freydan, 3: Chadegan, 4: Isfahan, 5: Mobarakeh, 6: Borkhar, 7: Khomeinishahr, 8: Shahinshahr, 9: Falavarjan, 10: Lenjan, 11: Dehaghan, 12: NajafAbad.</p> Full article ">Figure 1 Cont.
<p>Study area, (<b>a</b>) an overview of the study area: vegetated areas, the aquifer, irrigation networks, and the long-term mean annual precipitation sum (2003 to 2018). (<b>b</b>) Köppen–Geiger climate zones using the same color, BWh: Arid, desert, hot; BWk: Arid, desert, cold; BSk: Arid, steppe, cold; Dsa: Cold, dry summer, hot summer; Dsb: Cold, dry summer, warm summer; counties: 1: Tiran, 2: Freydan, 3: Chadegan, 4: Isfahan, 5: Mobarakeh, 6: Borkhar, 7: Khomeinishahr, 8: Shahinshahr, 9: Falavarjan, 10: Lenjan, 11: Dehaghan, 12: NajafAbad.</p> Full article ">Figure 2
<p>Workflow of ETa calculation. Dark blue cells represent products of this study. (<b>a</b>) ETa and scaled VIs calculation; cyan and green colors are representative of research questions a and b. (<b>b</b>) evaluation methods; light pink shows research questions c and d. VI: vegetation index, ET: evapotranspiration, EVI2<sub>MccL</sub> and EVI<sub>MccL</sub>: 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; EVI2<sub>ccL</sub> and EVI<sub>ccL</sub>: 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 3
<p>Maps and frequency graphs of VIs’ differences (<b>a</b>) 14 June 2002 and (<b>b</b>) 1 August 2008. EVI2<sub>MccL</sub> and EVI<sub>MccL</sub>: 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; EVI2<sub>ccL</sub> and EVI<sub>ccL</sub>: 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 4
<p>(<b>a</b>) Mean ETa estimated using the distinct VIs in millimeter considering static and dynamic harvested areas, (<b>b</b>) Total ETa estimated using the distinct VIs in cubic kilometer considering static and dynamic harvested areas, (<b>c</b>) red: annual cropland extent vs. blue: static harvested area (area harvested at least once). ET-EVI2<sub>MccL</sub> and ET-EVI<sub>MccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; ET-EVI2<sub>ccL</sub> and ET-EVI<sub>ccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 5
<p>Scaled VIs vs. FAO-Kc curves for (<b>a</b>) barley and (<b>b</b>) wheat. RMSE presents the magnitude of the errors in predictions and ranges from 0 to +∞ where smaller values show a better performance. The RMSE was applied to three quartiles (25, 50, and 75). EVI2<sub>MccL</sub> and EVI<sub>MccL</sub>: 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; EVI2<sub>ccL</sub> and EVI<sub>ccL</sub>: 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 6
<p>(<b>a</b>) Long-term mean ET-VI, and (<b>b</b>) differences between average of all ET-VIs and ET-VIs. ET-EVI2<sub>MccL</sub> and ET-EVI<sub>MccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; ET-EVI2<sub>ccL</sub> and ET-EVI<sub>ccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 7
<p>(<b>a</b>) Annual average ET-VIs over the ZRB, (<b>b</b>) Pixel-wise spatial correlation. ET-EVI2<sub>MccL</sub> and ET-EVI<sub>MccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; ET-EVI2<sub>ccL</sub> and ET-EVI<sub>ccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure 8
<p>Annual ETa anomaly boxplots. EVI2 based ETa anomalies tend to be slightly higher. ET-EVI2<sub>MccL</sub> and ET-EVI<sub>MccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (MODIS continuity-corrected Landsat) respectively; ET-EVI2<sub>ccL</sub> and ET-EVI<sub>ccL</sub>: ET calculated using 2- and 3-band Enhanced Vegetation Index (continuity-corrected Landsat) respectively.</p> Full article ">Figure A1
<p>(<b>a</b>) Frequency of the number of crop harvests in the 19-years, (<b>b</b>) Long-term MVC (maximum values composite), (<b>c</b>) Land use map of agricultural area in 2015 (prepared by Ministry of Energy of Iran). The red rectangle shows the area where no land use map was available.</p> Full article ">Figure A1 Cont.
<p>(<b>a</b>) Frequency of the number of crop harvests in the 19-years, (<b>b</b>) Long-term MVC (maximum values composite), (<b>c</b>) Land use map of agricultural area in 2015 (prepared by Ministry of Energy of Iran). The red rectangle shows the area where no land use map was available.</p> Full article ">Figure A2
<p>Vegetated areas (static cropland) and agriculture land use map (in blue). Zoom window shows an example of the area where in land use map expansion of croplands (green colors) cannot be seen. Values under 0.3 is soil and sparse vegetation.</p> Full article ">Figure A3
<p>Annual MVC (maximum values composite) map of 2015 and agricultural land use map.</p> Full article ">
19 pages, 3000 KiB  
Article
UAV- and Machine Learning-Based Retrieval of Wheat SPAD Values at the Overwintering Stage for Variety Screening
by Jianjun Wang, Qi Zhou, Jiali Shang, Chang Liu, Tingxuan Zhuang, Junjie Ding, Yunyu Xian, Lingtian Zhao, Weiling Wang, Guisheng Zhou, Changwei Tan and Zhongyang Huo
Remote Sens. 2021, 13(24), 5166; https://doi.org/10.3390/rs13245166 - 20 Dec 2021
Cited by 40 | Viewed by 4054
Abstract
In recent years, the delay in sowing has become a major obstacle to high wheat yield in Jiangsu Province, one of the major wheat producing areas in China; hence, it is necessary to screen wheat varieties are resilient for late sowing. This study [...] Read more.
In recent years, the delay in sowing has become a major obstacle to high wheat yield in Jiangsu Province, one of the major wheat producing areas in China; hence, it is necessary to screen wheat varieties are resilient for late sowing. This study aimed to provide an effective, fast, and non-destructive monitoring method of soil plant analysis development (SPAD) values, which can represent leaf chlorophyll contents, for late-sown winter wheat variety screening. This study acquired multispectral images using an unmanned aerial vehicle (UAV) at the overwintering stage of winter wheat growth, and further processed these images to extract reflectance of five single spectral bands and calculated 26 spectral vegetation indices. Based on these 31 variables, this study combined three variable selection methods (i.e., recursive feature elimination (RFE), random forest (RF), and Pearson correlation coefficient (r)) with four machine learning algorithms (i.e., random forest regression (RFR), linear kernel-based support vector regression (SVR), radial basis function (RBF) kernel-based SVR, and sigmoid kernel-based SVR), resulted in seven SVR models (i.e., RFE-SVR_linear, RF-SVR_linear, RF-SVR_RBF, RF-SVR_sigmoid, r-SVR_linear, r-SVR_RBF, and r-SVR_sigmoid) and three RFR models (i.e., RFE-RFR, RF-RFR, and r-RFR). The performances of the 10 machine learning models were evaluated and compared with each other according to the achieved coefficient of determination (R2), residual prediction deviation (RPD), root mean square error (RMSE), and relative RMSE (RRMSE) in SPAD estimation. Of the 10 models, the best one was the RF-SVR_sigmoid model, which was the combination of the RF variable selection method and the sigmoid kernel-based SVR algorithm. It achieved high accuracy in estimating SPAD values of the wheat canopy (R2 = 0.754, RPD = 2.017, RMSE = 1.716 and RRMSE = 4.504%). The newly developed UAV- and machine learning-based model provided a promising and real time method to monitor chlorophyll contents at the overwintering stage, which can benefit late-sown winter wheat variety screening. Full article
(This article belongs to the Special Issue Crop Growth Monitoring Using Remote Sensing: Progress, Challenges and Opportunities)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Map displays the location of the study area and spatial distribution of the 96 experimental plots. The RGB image was acquired on 13 January 2021 at the wheat overwintering stage using the multispectral imaging system of the DJI Phantom 4 Multispectral UAV.</p> Full article ">Figure 2
<p>Methodology workflow.</p> Full article ">Figure 3
<p>Absolute values of correlation coefficients (r) between the variables (<a href="#remotesensing-13-05166-t002" class="html-table">Table 2</a>) and SPAD values of wheat canopy. The orange colour represents the variables that included only the visible spectral bands, the green colour represents the variables that included the near-infrared spectral band but not the red edge spectral band, and the blue colour represents the variables that included the red edge spectral band.</p> Full article ">Figure 4
<p>Comparison of spectral curves for varying SPAD values. The SPAD and spectral reflectance values are averaged for the plots under four different rates of nitrogen fertilization (i.e., N0: 0 kg ha<sup>−1</sup>, N210: 210 kg ha<sup>−1</sup>, N270: 270 kg ha<sup>−1</sup>, and N330: 330 kg ha<sup>−1</sup> pure nitrogen), respectively. In general, reflectance increases with increasing SPAD, especially in the red edge and near-infrared regions.</p> Full article ">Figure 5
<p>Scatterplots between the measured SPAD and the estimated SPAD values estimated by the LOOCV, when the three models that achieved the best estimation performance were applied. The diagonal lines illustrate the 1:1 relation. (<b>a</b>) RFE-SVR_linear; (<b>b</b>) RF-SVR_sigmoid; (<b>c</b>) r-SVR_sigmoid.</p> Full article ">
17 pages, 6881 KiB  
Article
Towards the Sea Wind Measurement with the Airborne Scatterometer Having the Rotating-Beam Antenna Mounted over Fuselage
by Alexey Nekrasov and Alena Khachaturian
Remote Sens. 2021, 13(24), 5165; https://doi.org/10.3390/rs13245165 - 20 Dec 2021
Cited by 5 | Viewed by 2811
Abstract
Extension of the existing airborne radars’ applicability is a perspective approach to the remote sensing of the environment. Here we investigate the capability of the rotating-beam radar installed over the fuselage for the sea surface wind measurement based on the comparison of the [...] Read more.
Extension of the existing airborne radars’ applicability is a perspective approach to the remote sensing of the environment. Here we investigate the capability of the rotating-beam radar installed over the fuselage for the sea surface wind measurement based on the comparison of the backscatter with the respective geophysical model function (GMF). We also consider the robustness of the proposed approach to the partial shading of the underlying water surface by the aircraft nose, tail, and wings. The wind retrieval algorithms have been developed and evaluated using Monte-Carlo simulations. We find our results promising both for the development of new remote sensing systems as well as the functional enhancement of existing airborne radars. Full article
(This article belongs to the Special Issue Remote Sensing of Ocean Surface Winds)
►▼ Show Figures

Figure 1

Figure 1
<p>Azimuth sectors available for the underlying surface NRCS observation at the medium incidence angles and sectors shadowed by the aircraft’s nose, tail, and wings when the rotating-beam radar is mounted over the fuselage.</p> Full article ">Figure 2
<p>Simulation results in the case of the narrow nose, tail, and wings (<span class="html-italic">N</span> = 52) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 45° with 120 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 3
<p>Simulation results in the case of the medium nose, tail, and wings (<span class="html-italic">N</span> = 36) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 45° with 174 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 4
<p>Simulation results in the case of the wide nose, tail, and wings (<span class="html-italic">N</span> = 20) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 45° with 313 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 5
<p>Simulation results in the case of the narrow nose, tail, and wings (<span class="html-italic">N</span> = 52) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 60° with 120 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 6
<p>Simulation results in the case of the medium nose, tail, and wings (<span class="html-italic">N</span> = 36) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 60° with 174 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 7
<p>Simulation results in the case of the wide nose, tail, and wings (<span class="html-italic">N</span> = 20) with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 60° with 313 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure 8
<p>Comparative results of the maximum wind retrieval errors in the case of the whole 360° azimuthal NRCS curve available (<span class="html-italic">N</span> = 72), in the cases of the narrow, medium, and wide nose, tail, and wings (<span class="html-italic">N</span> = 52, 36, and 20, respectively), and in the case when only four azimuth sectors at the directions of 45°, 135°, 225°, and 315° are evadible: red asterisks are for the incidence angle of 45°, and green asterisks are for the incidence angle of 60°.</p> Full article ">Figure 9
<p>The maximum altitude dependence on the incidence angle.</p> Full article ">Figure A1
<p>Simulation results in the case of the observation <span class="html-italic">N</span> = 72 azimuth sectors at the directions of 0°, 5°, 10°, …, 355° relative to the aircraft course with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 45° with 87 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure A2
<p>Simulation results in the case of the observation <span class="html-italic">N</span> = 72 azimuth sectors at the directions of 0°, 5°, 10°, …, 355° relative to the aircraft course with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 60° with 87 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure A3
<p>Simulation results in the case of the observation <span class="html-italic">N</span> = 4 azimuth sectors at the directions of 45°, 135°, 225°, and 315° relative to the aircraft course with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 45° with 1565 integrated NRCS samples for each azimuth sector.</p> Full article ">Figure A4
<p>Simulation results in the case of the observation <span class="html-italic">N</span> = 4 azimuth sectors at the directions of 45°, 135°, 225°, and 315° relative to the aircraft course with an assumption of 0.2 dB instrumental noise at the wind speeds of 2–20 m/s for the incidence angle of 60° with 1565 integrated NRCS samples for each azimuth sector.</p> Full article ">
32 pages, 19467 KiB  
Article
A New Method (MINDED-BA) for Automatic Detection of Burned Areas Using Remote Sensing
by Eduardo R. Oliveira, Leonardo Disperati and Fátima L. Alves
Remote Sens. 2021, 13(24), 5164; https://doi.org/10.3390/rs13245164 - 20 Dec 2021
Cited by 6 | Viewed by 3365
Abstract
This work presents a change detection method (MINDED-BA) for determining burned extents from multispectral remote sensing imagery. It consists of a development of a previous model (MINDED), originally created to estimate flood extents, combining a multi-index image-differencing approach and the analysis of magnitudes [...] Read more.
This work presents a change detection method (MINDED-BA) for determining burned extents from multispectral remote sensing imagery. It consists of a development of a previous model (MINDED), originally created to estimate flood extents, combining a multi-index image-differencing approach and the analysis of magnitudes of the image-differencing statistics. The method was implemented, using Landsat and Sentinel-2 data, to estimate yearly burn extents within a study area located in northwest central Portugal, from 2000–2019. The modelling workflow includes several innovations, such as preprocessing steps to address some of the most important sources of error mentioned in the literature, and an optimal bin number selection procedure, the latter being the basis for the threshold selection for the classification of burn-related changes. The results of the model have been compared to an official yearly-burn-extent database and allow verifying the significant improvements introduced by both the pre-processing procedures and the multi-index approach. The high overall accuracies of the model (ca. 97%) and its levels of automatization (through open-source software) indicate potential for being a reliable method for systematic unsupervised classification of burned areas. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The Multi-Index Differencing method for Burned Areas (MINDED-BA) workflow (adapted from [<a href="#B34-remotesensing-13-05164" class="html-bibr">34</a>]).</p> Full article ">Figure 2
<p>Study area location map (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 3
<p>Surface reflectance signatures of different land cover types and their correspondence to Landsat 5 TM (adapted from [<a href="#B28-remotesensing-13-05164" class="html-bibr">28</a>]), along with different increments of a thresholding range (M1–M4) aimed for masking highly reflective surfaces (HRS).</p> Full article ">Figure 4
<p>Example of highly reflective surfaces (HRS) masks, obtained from different increments of the reference thresholding range (M1–M4 in <a href="#remotesensing-13-05164-f003" class="html-fig">Figure 3</a>), compared to the corresponding false color composite of a Landsat 8 OLI scene (from 06/11/2017) and the reference burned areas (RBA) (ICNF, 2020) (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 5
<p>Example of a Bin_ratio distribution of d1f, for each ΔBrI, in 2017.</p> Full article ">Figure 6
<p>Distribution of yearly optimal numbers of bins, per ΔBrI and derivative order: (<b>a</b>) <span class="html-italic">d1f</span>; (<b>b</b>) <span class="html-italic">d2f</span>.</p> Full article ">Figure 7
<p>Comparative example of different statistical data-binning intervals of ΔNBR2 from 2007, with the corresponding thresholds (×) for d1f (<b>a</b>) and d2f (<b>b</b>).</p> Full article ">Figure 8
<p>Distribution of the thresholds obtained for each ΔBrI: (<b>a</b>) T1; (<b>b</b>) T2 (if any).</p> Full article ">Figure 9
<p>Examples of ΔBrI coeval thematic maps, for 2006 and 2016, compared to the RBA [<a href="#B46-remotesensing-13-05164" class="html-bibr">46</a>] (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 10
<p>MINDED-BA maps from 2000 to 2019, in comparison to the RBA extents [<a href="#B46-remotesensing-13-05164" class="html-bibr">46</a>] (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 11
<p>False color RGB composites (R:SWIR2; G:SWIR1;B:NIR) of t0 and t1, together with MINDED-BA and RBA maps, for 2009 and 2016 (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 12
<p>Comparison of MINDED-BA maps obtained with Landsat 8 (LS8) and Sentinel-2A(S2A), alongside the RBA extents [<a href="#B46-remotesensing-13-05164" class="html-bibr">46</a>]: (<b>a</b>) 2018, based on LS8 only; (<b>b</b>) 2018-S, couple with LS8 and S2A; (<b>c</b>) overlay representation of (<b>a</b>,<b>b</b>); (<b>d</b>) close-up of (<b>c</b>), nearby waterbodies, and RBA; (<b>e</b>) 2019, based on LS8 only; (<b>f</b>) 2019-S, based on S2A only; (<b>g</b>) overlay representation of (<b>e</b>,<b>f</b>); (<b>h</b>) close-up of (<b>g</b>) and nearby waterbodies. Keys for classifications comparison: (i) matching classifications—same classes for both multitemporal analyses; (ii) mismatching classifications—different classifications between both analysis; (iii) other—combinations of classes including the ‘Mixed’ classification (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 13
<p>Comparison of area of clusters determined from the mismatching combinations related to the classifications 2018 vs. 2018-S and 2019 vs. 2019-S.</p> Full article ">Figure 14
<p>Summary of the confusion matrix statistics between MINDED-BA maps and the yearly reference burned areas (RBA) (Nc—no change; LMc—low-magnitude change; HMc—high-magnitude change; TC—total changes (LMc+HMc); Ov—overall accuracy).</p> Full article ">Figure 15
<p>MINDED-BA uncertainty maps from 2000 to 2019 (Coordinate System: PT-TM06/ETRS89).</p> Full article ">Figure 16
<p>Comparison of results obtained with different processing levels, including those from HRS masking and the multi-index majority analysis (example of the year 2017, with the single-index differencing corresponding to ΔNBR2) (Coordinate System: PT-TM06/ETRS89).</p> Full article ">
19 pages, 5189 KiB  
Article
A scSE-LinkNet Deep Learning Model for Daytime Sea Fog Detection
by Xiaofei Guo, Jianhua Wan, Shanwei Liu, Mingming Xu, Hui Sheng and Muhammad Yasir
Remote Sens. 2021, 13(24), 5163; https://doi.org/10.3390/rs13245163 - 20 Dec 2021
Cited by 13 | Viewed by 3684
Abstract
Sea fog is a precarious weather disaster affecting transportation on the sea. The accuracy of the threshold method for sea fog detection is limited by time and region. In comparison, the deep learning method learns features of objects through different network layers and [...] Read more.
Sea fog is a precarious weather disaster affecting transportation on the sea. The accuracy of the threshold method for sea fog detection is limited by time and region. In comparison, the deep learning method learns features of objects through different network layers and can therefore accurately extract fog data and is less affected by temporal and spatial factors. This study proposes a scSE-LinkNet model for daytime sea fog detection that leverages residual blocks to encoder feature maps and attention module to learn the features of sea fog data by considering spectral and spatial information of nodes. With the help of satellite radar data from Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), a ground sample database was extracted from Moderate Resolution Imaging Spectroradiometer (MODIS) L1B data. The scSE-LinkNet was trained on the training set, and quantitative evaluation was performed on the test set. Results showed the probability of detection (POD), false alarm rate (FAR), critical success index (CSI), and Heidke skill scores (HSS) were 0.924, 0.143, 0.800, and 0.864, respectively. Compared with other neural networks (FCN, U-Net, and LinkNet), the CSI of scSE-LinkNet was improved, with a maximum increase of nearly 8%. Moreover, the sea fog detection results were consistent with the measured data and CALIOP products. Full article
(This article belongs to the Special Issue Explainable Deep Neural Networks for Remote Sensing Image Understanding)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Geographical location of the study area.</p> Full article ">Figure 2
<p>Flow chart of the sea fog detection method.</p> Full article ">Figure 3
<p>LinkNet architecture: (<b>a</b>) the whole structure of LinkNet, (<b>b</b>) encoder block of LinkNet, and (<b>c</b>) decoder block of LinkNet. m represents input feature map; n represents output feature map.</p> Full article ">Figure 4
<p>SE module. <math display="inline"><semantics> <mrow> <msub> <mi>F</mi> <mrow> <mi>t</mi> <mi>r</mi> </mrow> </msub> <mo stretchy="false">(</mo> <mo>⋅</mo> <mo>,</mo> <mi>θ</mi> <mo stretchy="false">)</mo> </mrow> </semantics></math> represents any transformation of feature maps, <math display="inline"><semantics> <mrow> <msub> <mi>F</mi> <mrow> <mi>s</mi> <mi>q</mi> </mrow> </msub> <mo stretchy="false">(</mo> <mo>⋅</mo> <mo stretchy="false">)</mo> </mrow> </semantics></math> represents the squeeze transformation, <math display="inline"><semantics> <mrow> <msub> <mi>F</mi> <mrow> <mi>e</mi> <mi>x</mi> </mrow> </msub> <mo stretchy="false">(</mo> <mo>⋅</mo> <mo>,</mo> <mi>W</mi> <mo stretchy="false">)</mo> </mrow> </semantics></math> represents the excitation transformation, and <math display="inline"><semantics> <mrow> <msub> <mi>F</mi> <mrow> <mi>s</mi> <mi>c</mi> <mi>a</mi> <mi>l</mi> <mi>e</mi> </mrow> </msub> <mo stretchy="false">(</mo> <mo>⋅</mo> <mo>,</mo> <mo>⋅</mo> <mo stretchy="false">)</mo> </mrow> </semantics></math> represents the scale transformation.</p> Full article ">Figure 5
<p>The structure of scSE-LinkNet.</p> Full article ">Figure 6
<p>The structure of the spatial and channel squeeze-and-excitation block (scSE). σ(∙) represents sigmoid.</p> Full article ">Figure 7
<p>Convolutional modules in decoder block. m represents input feature map; n represents output feature map.</p> Full article ">Figure 8
<p>The superimposed result of MODIS and CALIOP data. (<b>a</b>) True color red–green–blue (RGB) images from 13 May 2017 at 05:00 UTC; the yellow line represents the CALIOP trajectory line, and the blue points represents sea fog points. (<b>b</b>) CALIOP VFM data profile. The sample points in red squares in (<b>a</b>,<b>b</b>) are sea fog samples.</p> Full article ">Figure 9
<p>The sea fog detection result of different deep learning model with samples from test set, (<b>a</b>–<b>e</b>) are randomly selected samples of the test set.</p> Full article ">Figure 10
<p>Distribution of the measured data in April 2018. The yellow points are measured data from meteorological station, and the purple points are measured data from ICOADS.</p> Full article ">Figure 11
<p>The sea fog detection results with different models. The white areas represents sea fog detection results, and the blue points represent the measured data consistent with sea fog detection results during the time of sea fog occurrence. (<b>a</b>) 1 April 2018 at 05:30 UTC, (<b>b</b>) 2 April 2018 at 04:35 UTC, (<b>c</b>) 30 April 2018 at 05:00 UTC.</p> Full article ">Figure 12
<p>The sea fog detection results and CALIOP VFM data profile at 05:00 UTC on 8 April 2016. (<b>a</b>) The MODIS image, (<b>b</b>) the verified points distribution in the sea fog detection result, (<b>c</b>) the VFM profile of the sea fog detection result.</p> Full article ">Figure 13
<p>Distribution of sea fog evaluation metrics with different sample numbers.</p> Full article ">
15 pages, 16223 KiB  
Article
A Complete Environmental Intelligence System for LiDAR-Based Vegetation Management in Power-Line Corridors
by Domen Mongus, Matej Brumen, Danijel Žlaus, Štefan Kohek, Roman Tomažič, Uroš Kerin and Simon Kolmanič
Remote Sens. 2021, 13(24), 5159; https://doi.org/10.3390/rs13245159 - 20 Dec 2021
Cited by 8 | Viewed by 3995
Abstract
This paper presents the first complete approach to achieving environmental intelligence support in the management of vegetation within electrical power transmission corridors. Contrary to the related studies that focused on the mapping of power lines, together with encroaching vegetation risk assessment, we realised [...] Read more.
This paper presents the first complete approach to achieving environmental intelligence support in the management of vegetation within electrical power transmission corridors. Contrary to the related studies that focused on the mapping of power lines, together with encroaching vegetation risk assessment, we realised predictive analytics with vegetation growth simulation. This was achieved by following the JDL/DFIG data fusion model for complementary feature extraction from Light Detection and Ranging (LiDAR) derived data products and auxiliary thematic maps that feed an ensemble regression model. The results indicate that improved vegetation growth prediction accuracy is obtained by segmenting training samples according to their contextual similarities that relate to their ecological niches. Furthermore, efficient situation assessment was then performed using a rasterised parametrically defined funnel-shaped volumetric filter. In this way, RMSE≈1 m was measured when considering tree growth simulation, while a 0.37 m error was estimated in encroaching vegetation detection, demonstrating significant improvements over the field observations. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Study area.</p> Full article ">Figure 2
<p>Overall concept of the data fusion framework.</p> Full article ">Figure 3
<p>Filter definition (<b>a</b>) using a parameterized funnel-shaped generator, swept along the power-transmission line axis in order to obtain (<b>b</b>) a raster layer of the maximum allowed vegetation heights.</p> Full article ">Figure 4
<p>Encroaching vegetation detection.</p> Full article ">Figure 5
<p>A flowchart of the proposed context-based ensemble regression.</p> Full article ">Figure 6
<p>Definition of regression features.</p> Full article ">Figure 7
<p>Distribution of measured errors achieved by the tested regression models with (C) and without (NC) using contextual segmentation of the learning data in comparison to measured tree growth.</p> Full article ">Figure 8
<p>Spatial distribution of errors, where (<b>a</b>) shows an input CHM, (<b>b</b>) ground-truth, (<b>c</b>) simulated CHM and (<b>d</b>) the difference between the latter, where blue colours indicate low values, while red colours are used to display high values.</p> Full article ">Figure 9
<p>Overlap between the areas of detected encroaching vegetation (red) and the clearance areas (green).</p> Full article ">Figure 10
<p>Over-detected encroaching vegetation (grey) outside of the clearance areas (green) with risk-areas (red).</p> Full article ">
22 pages, 3585 KiB  
Article
Mapping Soil Organic Matter and Analyzing the Prediction Accuracy of Typical Cropland Soil Types on the Northern Songnen Plain
by Meiwei Zhang, Huanjun Liu, Meinan Zhang, Haoxuan Yang, Yuanliang Jin, Yu Han, Haitao Tang, Xiaohan Zhang and Xinle Zhang
Remote Sens. 2021, 13(24), 5162; https://doi.org/10.3390/rs13245162 - 19 Dec 2021
Cited by 14 | Viewed by 3562
Abstract
Soil organic matter (SOM) plays a critical role in agroecosystems and the terrestrial carbon cycle. Thus, accurately mapping SOM promotes sustainable agriculture and estimations of soil carbon pools. However, few studies have analyzed the changing trends in multi-period SOM prediction accuracies for single [...] Read more.
Soil organic matter (SOM) plays a critical role in agroecosystems and the terrestrial carbon cycle. Thus, accurately mapping SOM promotes sustainable agriculture and estimations of soil carbon pools. However, few studies have analyzed the changing trends in multi-period SOM prediction accuracies for single cropland soil types and mapped their spatial SOM patterns. Using time series 7 MOD09A1 images during the bare soil period, we combined the pixel dates of training samples and precipitation data to explore the variation in SOM accuracy for two typical cropland soil types. The advantage of using single soil type data versus the total dataset was evaluated, and SOM maps were drawn for the northern Songnen Plain. When almost no precipitation occurred on or near the optimal pixel date, the accuracies increased, and vice versa. SOM models of the two soil types achieved a lower root mean squared error (RMSE = 0.55%, 0.79%) and mean absolute error (MAE = 0.39%, 0.58%) and a higher coefficient of determination (R2 = 0.65, 0.75) than the model using the total dataset and resulted in a mean relative improvement (RI) of 30.21%. The SOM decreased from northeast to southwest. The results provide reference data for the accurate management of cultivated soil and determining carbon sequestration. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Workflow schematic for analyzing the prediction accuracy of regional soil organic matter (SOM) and mapping its spatial distribution. Notably, the models using two single soil types data (i.e., Arenosols and Phaeozems) are called the “two typical soil types models” and using the total dataset are called the “total dataset models”.</p> Full article ">Figure 2
<p>Overview of the northern Songnen Plain and study area (<b>a</b>), soil sampling locations of training samples, and main meteorological stations for Arenosols and Phaeozems (<b>b</b>). Photograph of the soil surface condition after plowing (<b>c</b>).</p> Full article ">Figure 3
<p>Time series of precipitation and root mean squared error (<span class="html-italic">RMSE</span>) on different days of the year (DOYs) for Phaeozems (<b>a</b>) and Arenosols (<b>b</b>). The <span class="html-italic">x</span>-axis values corresponding to the red dots represent the optimal pixel dates of training samples based on 7 images. The histogram values denote the precipitation on different dates. Notably, a smaller <span class="html-italic">RMSE</span> value indicates a higher prediction accuracy, whereas a larger <span class="html-italic">RMSE</span> value indicates a lower prediction accuracy.</p> Full article ">Figure 4
<p>Spatial distribution patterns of residual values for the total (<b>a</b>), Arenosol (<b>b</b>), and Phaeozem (<b>c</b>) sample datasets.</p> Full article ">Figure 5
<p>Maps of SOM content in the study area using the total dataset (<b>a</b>) and two typical cropland soil types (i.e., Arenosols and Phaeozems) (<b>b</b>).</p> Full article ">
14 pages, 2758 KiB  
Article
Radiative Transfer Model Simulations for Ground-Based Microwave Radiometers in North China
by Wenying He, Yunchu Cheng, Rongshi Zou, Pucai Wang, Hongbin Chen, Jun Li and Xiangao Xia
Remote Sens. 2021, 13(24), 5161; https://doi.org/10.3390/rs13245161 - 19 Dec 2021
Cited by 3 | Viewed by 2758
Abstract
Ground-based microwave radiometer profilers (MWRPs) are widely used to provide high-temporal resolution atmospheric temperature and humidity profiles. The quality of the observed brightness temperature (TB) from MWRPs is key for retrieving accurate atmospheric profiles. In this study, TB simulations derived from a radiative [...] Read more.
Ground-based microwave radiometer profilers (MWRPs) are widely used to provide high-temporal resolution atmospheric temperature and humidity profiles. The quality of the observed brightness temperature (TB) from MWRPs is key for retrieving accurate atmospheric profiles. In this study, TB simulations derived from a radiative transfer model (RTM) were used to assess the quality of TB observations. Two types of atmospheric profile data (conventional radiosonde and ERA5 reanalysis) were combined with the RTM to obtain TB simulations, then compared with corresponding observations from three MWRPs located in different places in North China to investigate the influence of input atmospheric profiles on TB simulations and evaluate the quality of TB observations from the three MWRPs. The comparisons of the matching samples under clear-sky conditions showed that TB simulations derived from both radiosonde and ERA5 profiles were very close to the TB observations from most of the MWRP channels; however, the correlation was lower and the bias was obvious at 51.26 GHz and 52.28 GHz, which indicates that the oxygen absorption component in the RTM needs to be improved for lower-frequency temperature channels. The difference in location of the radiosonde and MWRP sites affected the TB simulations for the water vapor channels, but had little impact on temperature channels that are insensitive to humidity. Comparisons of both simulations (ERA5 and Radiosonde) and the corresponding TB observations from the three sites indicated that the water vapor channels observation quality for the MWRP located in southern Beijing needs improvement. For the two types of profile data, ERA5 profiles have a more positive effect on TB simulations in the water vapor channels, such as enhanced consistence, reduced bias and standard deviation between simulations and observations for those MWRPs located away from the radiosonde station. Therefore, hourly ERA5 data are an optimal option in terms of compensating for limited radiosonde measurements and enhancing the monitoring quality of MWRP observations within 24 h. Full article
(This article belongs to the Special Issue Feature Papers of Section Atmosphere Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>The ground-based MWRPs used (<b>a</b>) and the location of three sites, as indicated by the red markers on the map (<b>b</b>). IAP is approximately 20 km from GXT and XH is approximately 42 km from GXT. Beijing radiosonde (No.54511) is launched from the GXT site.</p> Full article ">Figure 2
<p>The scattering plots of the simulated TBs using radiosonde profiles and observed TBs in channel 1 (Ch1), Ch7, and Ch8 of the MWRPs at the GXT (<b>a</b>–<b>c</b>), IAP (<b>d</b>–<b>f</b>), and XH (<b>g</b>–<b>i</b>) sites in 2019.</p> Full article ">Figure 3
<p>Comparisons of TB difference ((<b>a</b>) <span class="html-italic">CC</span>; (<b>b</b>) <span class="html-italic">MB</span>; (<b>c</b>) <span class="html-italic">STD</span>) between the observations and simulations using the Beijing radiosonde profiles at the three sites.</p> Full article ">Figure 4
<p>Comparisons of TB difference between the observations and simulations using the ERA5 profiles in Ch1, Ch7, and Ch8 of the MWRPs at the GXT (<b>a</b>–<b>c</b>), IAP (<b>d</b>–<b>f</b>), and XH (<b>g</b>–<b>i</b>) sites in 2019.</p> Full article ">Figure 5
<p>Comparisons of the statistical parameters ((<b>a</b>) <span class="html-italic">CC</span>; (<b>b</b>) <span class="html-italic">MB</span>; (<b>c</b>) <span class="html-italic">STD</span>) of the simulated TBs derived from both the radiosonde and ERA5 profiles with corresponding observed TBs in the water vapor channels of the MWRPs located at the GXT, IAP, and XH sites.</p> Full article ">Figure 6
<p>Monthly mean statistical parameters, <span class="html-italic">MB</span> (<b>a</b>) and <span class="html-italic">STD</span> (<b>b</b>), for the TB difference between the simulated TBs derived from ERA5 profiles and the observed TBs in Ch 7 of the MWRPs located at the GXT, IAP, and XH sites in 2019.</p> Full article ">
18 pages, 4998 KiB  
Article
Assessing Post-Fire Effects on Soil Loss Combining Burn Severity and Advanced Erosion Modeling in Malesina, Central Greece
by Ioanna Tselka, Pavlos Krassakis, Alkiviadis Rentzelos, Nikolaos Koukouzas and Issaak Parcharidis
Remote Sens. 2021, 13(24), 5160; https://doi.org/10.3390/rs13245160 - 19 Dec 2021
Cited by 10 | Viewed by 4666
Abstract
Earth’s ecosystems are extremely valuable to humanity, playing a key role ecologically, economically, and socially. Wildfires constitute a significant threat to the environment, especially in vulnerable ecosystems, such as those that are commonly found in the Mediterranean. Due to their strong impact on [...] Read more.
Earth’s ecosystems are extremely valuable to humanity, playing a key role ecologically, economically, and socially. Wildfires constitute a significant threat to the environment, especially in vulnerable ecosystems, such as those that are commonly found in the Mediterranean. Due to their strong impact on the environment, they provide a crucial factor in managing ecosystems behavior, causing dramatic modifications to land surface processes dynamics leading to land degradation. The soil erosion phenomenon downgrades soil quality in ecosystems and reduces land productivity. Thus, it is imperative to implement advanced erosion prediction models to assess fire effects on soil characteristics. This study focuses on examining the wildfire case that burned 30 km2 in Malesina of Central Greece in 2014. The added value of remote sensing today, such as the high accuracy of satellite data, has contributed to visualizing the burned area concerning the severity of the event. Additional data from local weather stations were used to quantify soil loss on a seasonal basis using RUSLE modeling before and after the wildfire. Results of this study revealed that there is a remarkable variety of high soil loss values, especially in winter periods. More particularly, there was a 30% soil loss rise one year after the wildfire, while five years after the event, an almost double reduction was observed. In specific areas with high soil erosion values, infrastructure works were carried out validating the applied methodology. The approach adopted in this study underlines the significance of using remote sensing and geoinformation techniques to assess the post-fire effects of identifying vulnerable areas based on soil erosion parameters on a local scale. Full article
(This article belongs to the Special Issue Integrated Use of Earth Observation and GIS Approaches for Soil Erosion Assessment in Local, Regional and Global Scale)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Location of the study area.</p> Full article ">Figure 2
<p>Geomorphological map of the study area. Red dashed polygon represents the boundaries of the affected area [<a href="#B22-remotesensing-13-05160" class="html-bibr">22</a>].</p> Full article ">Figure 3
<p>Land cover of the burned area according to Corine Land Cover 2012 [<a href="#B23-remotesensing-13-05160" class="html-bibr">23</a>].</p> Full article ">Figure 4
<p>Infrared color Landsat 8 images before (<b>left</b>) and after (<b>right</b>) the fire event.</p> Full article ">Figure 5
<p>Burned areas classified according to burn severity of the study area.</p> Full article ">Figure 6
<p>Seasonal spatial distribution of soil erosion rate for Mazi of Malesina in 2013 and 2015.</p> Full article ">Figure 7
<p>Seasonal spatial distribution of soil erosion rate for Mazi of Malesina in 2015 and 2020.</p> Full article ">Figure 8
<p>Burn severity quantification and land cover classification.</p> Full article ">Figure 9
<p>Seasonal variation of mean soil loss rate in 2013, 2015, and 2020.</p> Full article ">Figure 10
<p>Google Earth imagery (obtained in 2021) shows depicts areas where engineering infrastructure works have been recently implemented. Yellow-shaped polygon represents the boundaries of the burned area.</p> Full article ">
18 pages, 5772 KiB  
Article
Retrieval of Daily Mean VIIRS SST Products in China Seas
by Qianmei Li, Qingyou He and Chuqun Chen
Remote Sens. 2021, 13(24), 5158; https://doi.org/10.3390/rs13245158 - 19 Dec 2021
Cited by 1 | Viewed by 2621
Abstract
Sea surface temperature (SST) is one of the most important factors in regulating air-sea heat flux and, thus, climate change. Most of current global daily SST products are derived from one or two transient measurements of polar-orbiting satellites, which are not the same [...] Read more.
Sea surface temperature (SST) is one of the most important factors in regulating air-sea heat flux and, thus, climate change. Most of current global daily SST products are derived from one or two transient measurements of polar-orbiting satellites, which are not the same to daily mean SST values. In this study, high-temporal-resolution SST measurements (32–40 snapshots per day) from a geostationary satellite, FengYun-4A (FY–4A), are used to analyze the diurnal variation of SST in China seas. The results present a sinusoidal pattern of the diurnal variability in SST, with the maximum value at 13:00–15:00 CST and the minimum at 06:00–08:00 CST. Based on the diurnal variation of SST, a retrieval method for daily mean SST products from polar-orbiting satellites is established and applied to 7716 visible infrared imaging radiometer (VIIRS) data in China seas. The results suggest that it is feasible and practical for the retrieval of daily mean SST with an average RMSE of 0.133 °C. This retrieval method can also be utilized to other polar-orbiting satellites and obtain more daily mean satellite SST products, which will contribute to more accurate estimation and prediction between atmosphere and ocean in the future. Full article
(This article belongs to the Special Issue VIIRS 2011–2021: Ten Years of Success in Earth Observations)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Satellite zenith distribution of FY–4A/AGRI scanning range to the earth (Gray filled area is land mask. Bold red dashed frame represents sea area studied in this paper). Zone 1 to Zone 3 are used in the discussion of diurnal variation of SST in different zones in the latter part of this paper.</p> Full article ">Figure 2
<p>Distribution of FY–4A/AGRI matchups with (<b>a</b>) in-situ measurements and (<b>b</b>) S-NPP/VIIRS respectively, during June 2018 to May 2019 in China seas. The colors are the number of matchups in 1° × 1° resolution. Frequency diagrams (<b>c</b>,<b>d</b>) plotting SST comparison of matchups in figures (<b>a</b>,<b>b</b>), respectively. The colors denote the total number of SST values in each 0.5 °C × 0.5 °C box.</p> Full article ">Figure 3
<p>SST diurnal variation (SST<sub>dv</sub>) at different hours on 9 February 2019 in the study area. The colors indicate the difference (°C) between instantaneous SST and the daily mean SST. These SST data are from level–2 products of FY–4A/AGRI. The blank areas in the figure are missing data due to cloud pollution.</p> Full article ">Figure 4
<p>(<b>a</b>) The monthly mean SST (°C) and (<b>b</b>) the monthly mean values of SST<sub>amp_max</sub> in bins of 1° × 1° (°C) during October 2018 in China seas. The SST<sub>amp_max</sub> is the maximum difference of SST in a day. Time series of SST diurnal variation in area X<sub>1</sub> (<b>c</b>) and area X<sub>2</sub> (<b>d</b>) in October 2018. The blue dots are monthly mean values of SST (°C) for each hour and the black line is the trendline obtained by smoothing the scatters.</p> Full article ">Figure 5
<p>Time series of monthly mean SST (<b>a</b>–<b>c</b>) and monthly mean SST<sub>dv</sub> (<b>d</b>–<b>f</b>) in January, April, July and October (representing winter, spring, summer and autumn, respectively) for the three zones. SST<sub>dv</sub> (°C) = SST<sub>i</sub>–SST<sub>daily_mean</sub>, where i represents different hours. Zone 1 includes 0–15°N and Zone 2 includes 15–30°N, while Zone 3 includes 30–45°N.</p> Full article ">Figure 6
<p>(<b>a</b>–<b>d</b>) Time series comparison of monthly mean SST<sub>dv</sub> (°C) for the three zones (January, April, July and October represent winter, spring, summer and autumn, respectively). It shows the diurnal range of SST and the time of extreme SST in different zones during the same month.</p> Full article ">Figure 7
<p>Diurnal variation of AGRI SST (blue dots) and VIIRS SST (red triangles) in Zone 2 on 25 June 2018. VIIRS SST was obtained by the sensor aboard S-NPP at approximately 13:30 p.m. (ascending node). The solid black line is time series curve of AGRI SST for the day, which is smoothed from the scatters.</p> Full article ">Figure 8
<p>(<b>a</b>–<b>d</b>) Validation for retrieval of daily mean SST in the study area, while the spatiotemporal consistent daily means of AGRI SST are applied for comparison with the daytime VIIRS SST (<b>left panels</b>) and retrieved daily mean VIIRS SST (<b>right panels</b>) on October 2018 and May 2019.</p> Full article ">Figure 9
<p>(<b>a</b>–<b>f</b>) Validation for retrieved results of daily mean SST in the three zones from June 2018 to May 2019, while the spatiotemporal consistent daily means of AGRI SST are applied for comparison with the daytime VIIRS SST (<b>left panels</b>) and retrieved daily mean VIIRS SST (<b>right panels</b>).</p> Full article ">Figure 10
<p>(<b>a</b>) Time series curves of daily mean AGRI SST (solid red curve), instantaneous SST of VIIRS during daytime (dashed blue curve), retrieved VIIRS SST (solid blue curve) and the mean of instantaneous VIIRS SST between day and night (dashed pink curve), during July 2018. (<b>b</b>) The difference between the other three kinds of SST in (<b>a</b>) and the daily mean of AGRI SST.</p> Full article ">Figure A1
<p>Distribution diagrams of peak solar radiation (<b>a</b>–<b>d</b>) and wind speed at 10 m surface (<b>e</b>–<b>h</b>) during different seasons in the study area (January, April, July and October represent winter, spring, summer and autumn respectively). These data are from the NCEP Climate Forecast System Version 2.</p> Full article ">
22 pages, 14289 KiB  
Article
Feasibility of Ground-Based Sky-Camera HDR Imagery to Determine Solar Irradiance and Sky Radiance over Different Geometries and Sky Conditions
by Pedro C. Valdelomar, José L. Gómez-Amo, Caterina Peris-Ferrús, Francesco Scarlatti and María Pilar Utrillas
Remote Sens. 2021, 13(24), 5157; https://doi.org/10.3390/rs13245157 - 19 Dec 2021
Cited by 12 | Viewed by 3702
Abstract
We propose a methodological approach to provide the accurate and calibrated measurements of sky radiance and broadband solar irradiance using the High Dynamic Range (HDR) images of a sky-camera. This approach is based on a detailed instrumental characterization of a SONA sky-camera in [...] Read more.
We propose a methodological approach to provide the accurate and calibrated measurements of sky radiance and broadband solar irradiance using the High Dynamic Range (HDR) images of a sky-camera. This approach is based on a detailed instrumental characterization of a SONA sky-camera in terms of image acquisition and processing, as well as geometric and radiometric calibrations. As a result, a 1 min time resolution database of geometrically and radiometrically calibrated HDR images has been created and has been available since February 2020, with daily updates. An extensive validation of our radiometric retrievals has been performed in all sky conditions. Our results show a very good agreement with the independent measurements of the AERONET almucantar for sky radiance and pyranometers for broadband retrievals. The SONA sky radiance shows a difference of an RMBD < 10% while the broadband diffuse radiation shows differences of 2% and 5% over a horizontal plane and arbitrarily oriented surfaces, respectively. These results support the developed methodology and allow us to glimpse the great potential of sky-cameras to carry out accurate measurements of sky radiance and solar radiation components. Thus, the remote sensing techniques described here will undoubtedly be of great help for solar and atmospheric research. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Sky dome projection on our SONA sky-camera: (<b>a</b>) Pixel Zenith Angle; (<b>b</b>) Pixel Azimuth Angle; (<b>c</b>) Pixel Solid Angle.</p> Full article ">Figure 2
<p>Example of the determination to obtain the response function of the sensor for the blue channel: (<b>a</b>) fragments of the characteristic curve at nine exposition times; (<b>b</b>) the fitting characteristic curve.</p> Full article ">Figure 3
<p>Comparison of weighting functions. Original triangular function [<a href="#B41-remotesensing-13-05157" class="html-bibr">41</a>] and our own function <span class="html-italic">w</span>(<span class="html-italic">z</span>), with <span class="html-italic">b</span> = 8 and <span class="html-italic">a</span> = 12 and 24.</p> Full article ">Figure 4
<p>Sky-camera response function obtained for the three channels (RGB) using the weighting function in Equation (2).</p> Full article ">Figure 5
<p>Normalized Spectral Response of BFLY-PGE-20E4C-CS Point Grey Camera for RGB channels. (From Point Gray Imaging Performance Specification v7.0).</p> Full article ">Figure 6
<p>Probability density function of the calibration factor for each channel at its effective wavelength: 480 nm, 541 nm and 615 nm for the blue, green and red channels, respectively.</p> Full article ">Figure 7
<p>Examples of the representation of the almucantar (red dots) and principal (green dots) planes of the sun in an image on 27 June 2021 at 07:30. Isolines of different scattering angles are also shown with a 10° step.</p> Full article ">Figure 8
<p>(<b>a</b>) Mask to obtain the DHI. (<b>b</b>) Red calibrated Irradiance in µW cm<sup>−2</sup>. (<b>c</b>) Green calibrated Irradiance in µW cm<sup>−2</sup>. (<b>d</b>) Blue calibrated irradiance in µW cm<sup>−2</sup>. From a calibrated HDR sky-camera image on 27 June 2021 at 07:30 h under clear sky conditions.</p> Full article ">Figure 9
<p>Sketch of an arbitrary tilted and oriented plane and the transposition of the coordinated reference system.</p> Full article ">Figure 10
<p>Example of a PZA’ matrix for a tilted plane of 30° and oriented to geographical South with an azimuthal misalignment of 4.58°.</p> Full article ">Figure 11
<p>Comparison of pyranometer DHI vs. SONA-DHI from 12 February 2020 to 26 March 2021: (<b>a</b>) scatter plot, not considering saturated pixels; (<b>b</b>) scatter plot including saturated pixels; (<b>c</b>) residuals, not considering saturated pixels; and (<b>d</b>) residuals including the saturated pixels. The color bar accounts for the percentage of saturated pixels in each image, while the grey color represents the images with no saturated pixels. f(x) is the fit function for all images, represented by the red colored line, and g(x) is the fit line using only images without saturated pixels, represented by the green colored line. Two red dashed lines represent the 95% confidence interval of the residuals. Note that the residual is defined as the pyranometer DHI minus SONA DHI.</p> Full article ">Figure 12
<p>Comparison of the pyranometer DHI vs. SONA DHI from 28 March to 27 July 2021. SONA DHI were computed by applying the spectral response correction obtained in the comparison against the pyranometer, treating the saturated pixels as the maximum radiance. The color bar accounts for the percentage of saturated pixels in each image, while the grey color represents the images with no saturated pixels. f(x) is the fit function for all the images, represented by the red colored line, and g(x) is the fit line using only images without saturated pixels, represented by the green colored line. Two red dashed lines represent the 95% confidence interval of the residuals.</p> Full article ">Figure 13
<p>Comparison of the Pyranometer DHI on a 30° tilted plane oriented to South and the DHI retrieval from 4 to 5 September 2021. The samples are gradated according to their percentage of saturated pixels. f(x) is the fit function, represented by the red colored line. Two red dashed lines represent the 95% confidence interval of the residuals.</p> Full article ">
21 pages, 13759 KiB  
Article
Artificial Neural Network-Based Microwave Satellite Soil Moisture Reconstruction over the Qinghai–Tibet Plateau, China
by Jie Wang and Duanyang Xu
Remote Sens. 2021, 13(24), 5156; https://doi.org/10.3390/rs13245156 - 19 Dec 2021
Cited by 5 | Viewed by 2967
Abstract
Soil moisture is a key parameter for land-atmosphere interaction system; however, fewer existing spatial-temporally continuous and high-quality observation records impose great limitations on the application of soil moisture on long term climate change monitoring and predicting. Therefore, this study selected the Qinghai–Tibet Plateau [...] Read more.
Soil moisture is a key parameter for land-atmosphere interaction system; however, fewer existing spatial-temporally continuous and high-quality observation records impose great limitations on the application of soil moisture on long term climate change monitoring and predicting. Therefore, this study selected the Qinghai–Tibet Plateau (QTP) of China as research region, and explored the feasibility of using Artificial Neural Network (ANN) to reconstruct soil moisture product based on AMSR-2/AMSR-E brightness temperature and SMAP satellite data by introducing auxiliary variables, specifically considering the sensitivity of different combination of input variables, number of neurons in hidden layer, sample ratio, and precipitation threshold in model building. The results showed that the ANN model had the highest accuracy when all variables were used as inputs, it had a network containing 12 neurons in a hidden layer, it had a sample ratio 80%-10%-10% (training-validation-testing), and had a precipitation threshold of 8.75 mm, respectively. Furthermore, validation of the reconstructed soil moisture product (named ANN-SM) in other period were conducted by comparing with SMAP (April 2019 to July 2021) for all grid cells and in situ soil moisture sites (August 2010 to March 2015) of QTP, which achieved an ideal accuracy. In general, the proposed method is capable of rebuilding soil moisture products by adopting different satellite data and our soil moisture product is promising for serving the studies of long-term global and regional dynamics in water cycle and climate. Full article
(This article belongs to the Special Issue Data Science and Machine Learning for Geodetic Earth Observation)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>The location of Qinghai–Tibet Plateau (<b>a</b>) and distribution of in situ sites with LAI map. (<b>b</b>–<b>d</b>) corresponds to the Ngari network, the Naqu network, and the Maqu network, respectively.</p> Full article ">Figure 2
<p>Flowchart of the training, simulation, and validation process of the artificial neural network soil moisture (ANN-SM).</p> Full article ">Figure 3
<p>Mean of RMSE and R of all grid cells in the QTP for training, validation, testing, and overall samples in the context of sensitivity analysis in terms of (<b>a</b>) the combination of different input variables, (<b>b</b>) the number of hidden neurons, (<b>c</b>) different sample ratios, and (<b>d</b>) different value of precipitation thresholds. The purple box is the optimal parameter.</p> Full article ">Figure 4
<p>Error distribution maps of ANN-SM vs. SMAP soil moisture during the test period (April 2019 to March 2021) in terms of (<b>a</b>) R, (<b>b</b>) RMSE, and (<b>c</b>) BIAS.</p> Full article ">Figure 5
<p>Temporal series scatterplots for ANN-SM (purple hollow circles), LPRMsm (purple dots), JAXAsm (green cross), and SMOSsm (blue dots) vs. in situ soil moisture (black solid line) during a time period (August 2010 to March 2015) of (<b>a</b>) Maqu1 grid, (<b>b</b>) Maqu3 grid, (<b>c</b>) Naqu1 grid, (<b>d</b>) Naqu5 grid, (<b>e</b>) Ngari1 grid, and (<b>f</b>) Ngari2 grid.</p> Full article ">Figure 6
<p>The scatterplots for ANN-SM (x axis) vs. in situ soil moisture (y axis) during a time period (August 2010 to March 2015) of all grids.</p> Full article ">Figure A1
<p>Flowchart showing the training process of the ANN method.</p> Full article ">Figure A2
<p>Temporal series scatterplots for ANN-SM (purple hollow circles), LPRMsm (purple dots), JAXAsm (green cross), and SMOSsm (blue dots) vs. in situ soil moisture (black solid line) during a time period (August 2010 to March 2015) of (<b>a</b>) Maqu2 grid, (<b>b</b>) Maqu4 grid, (<b>c</b>) Maqu5 grid, (<b>d</b>) Maqu6 grid, (<b>e</b>) Maqu7 grid, (<b>f</b>) Maqu8 grid, (<b>g</b>) Naqu2 grid, (<b>h</b>) Naqu3 grid, and (<b>i</b>) Naqu4 grid.</p> Full article ">Figure A3
<p>The number of valid days retained after data removal through different precipitation thresholds. The number of days reserved when the precipitation threshold is 0 mm (<b>a</b>), and the number of further reduced days when the precipitation threshold is (<b>b</b>) 17.5 mm, (<b>c</b>) 12.5 mm, and (<b>d</b>) 8.75 mm.</p> Full article ">
18 pages, 4492 KiB  
Article
Modeling Influence of Soil Properties in Different Gradients of Soil Moisture: The Case of the Valencia Anchor Station Validation Site, Spain
by Ester Carbó, Pablo Juan, Carlos Añó, Somnath Chaudhuri, Carlos Diaz-Avalos and Ernesto López-Baeza
Remote Sens. 2021, 13(24), 5155; https://doi.org/10.3390/rs13245155 - 19 Dec 2021
Cited by 3 | Viewed by 3339
Abstract
The prediction of spatial and temporal variation of soil water content brings numerous benefits in the studies of soil. However, it requires a considerable number of covariates to be included in the study, complicating the analysis. Integrated nested Laplace approximations (INLA) with stochastic [...] Read more.
The prediction of spatial and temporal variation of soil water content brings numerous benefits in the studies of soil. However, it requires a considerable number of covariates to be included in the study, complicating the analysis. Integrated nested Laplace approximations (INLA) with stochastic partial differential equation (SPDE) methodology is a possible approach that allows the inclusion of covariates in an easy way. The current study has been conducted using INLA-SPDE to study soil moisture in the area of the Valencia Anchor Station (VAS), soil moisture validation site for the European Space Agency SMOS (Soil Moisture and Ocean Salinity). The data used were collected in a typical ecosystem of the semiarid Mediterranean conditions, subdivided into physio-hydrological units (SMOS units) which presents a certain degree of internal uniformity with respect to hydrological parameters and capture the spatial and temporal variation of soil moisture at the local fine scale. The paper advances the knowledge of the influence of hydrodynamic properties on VAS soil moisture (texture, porosity/bulk density and soil organic matter and land use). With the goal of understanding the factors that affect the variability of soil moisture in the SMOS pixel (50 km × 50 km), five states of soil moisture are proposed. We observed that the model with all covariates and spatial effect has the lowest DIC value. In addition, the correlation coefficient was close to 1 for the relationship between observed and predicted values. The methodology applied presents the possibility to analyze the significance of different covariates having spatial and temporal effects. This process is substantially faster and more effective than traditional kriging. The findings of this study demonstrate an advancement in that framework, demonstrating that it is faster than previous methodologies, provides significance of individual covariates, is reproducible, and is easy to compare with models. Full article
(This article belongs to the Special Issue Earth Observation in Support of Sustainable Soils Development)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study area with detailed physio-hydrological units. Area of the field campaigns highlighted in yellow (2008), brown (2009) and green (2010).</p> Full article ">Figure 2
<p>Density plot of original data. Theta probe mV mean (mV), Volumetric water content mean (%), Gravimetric water content mean (%), Laboratory bulk density mean (g cm<sup>−3</sup>), Cylinder bulk density mean (g cm<sup>−3</sup>), Cylinder plot bulk density mean (g cm<sup>−3</sup>), Cylinder plot porosity mean (%), Soil organic matter (%), Sand (%), Silt (%), Clay (%).</p> Full article ">Figure 3
<p>Density plot of variables used in designing the models (selected variables).</p> Full article ">Figure 4
<p>Mutual correlation values among response variable and covariates used in the models.</p> Full article ">Figure 5
<p>Proposal of soil moisture (%) distribution for a SMOS pixel under maximum (left) and minimum (right) soil moisture conditions. Unit in % volume/volume represents volume of water per total volume of soil.</p> Full article ">Figure 6
<p>Stochastic Partial Differential Equation (SPDE) meshes for spatial study. Top line, from left to right every study day, day 1 and day 2. Bottom line, from left to right days 3 to 5. The sample points are highlighted in green.</p> Full article ">Figure 7
<p>Scatter plot of model 2, observed vs. predicted (soil moisture content in %).</p> Full article ">Figure 8
<p>Prediction maps from model 1 (<b>a</b>) to model 5 (<b>e</b>). depicting soil moisture content in %.</p> Full article ">Figure 9
<p>Prediction maps for Soil Moisture states (SM 1 to 3) for model 1 (left) model 2 (right).</p> Full article ">
31 pages, 13635 KiB  
Article
Flood Risk Assessment of Metro System Using Improved Trapezoidal Fuzzy AHP: A Case Study of Guangzhou
by Guangpeng Wang, Lianyou Liu, Peijun Shi, Guoming Zhang and Jifu Liu
Remote Sens. 2021, 13(24), 5154; https://doi.org/10.3390/rs13245154 - 18 Dec 2021
Cited by 31 | Viewed by 5486
Abstract
Metro systems have become high-risk entities due to the increased frequency and severity of urban flooding. Therefore, understanding the flood risk of metro systems is a prerequisite for mega-cities’ flood protection and risk management. This study proposes a method for accurately assessing the [...] Read more.
Metro systems have become high-risk entities due to the increased frequency and severity of urban flooding. Therefore, understanding the flood risk of metro systems is a prerequisite for mega-cities’ flood protection and risk management. This study proposes a method for accurately assessing the flood risk of metro systems based on an improved trapezoidal fuzzy analytic hierarchy process (AHP). We applied this method to assess the flood risk of 14 lines and 268 stations of the Guangzhou Metro. The risk results validation showed that the accuracy of the improved trapezoidal fuzzy AHP (90% match) outperformed the traditional trapezoidal AHP (70% match). The distribution of different flood risk levels in Guangzhou metro lines exhibited a polarization signature. About 69% (155 km2) of very high and high risk zones were concentrated in central urban areas (Yuexiu, Liwan, Tianhe, and Haizhu); the three metro lines with the highest overall risk level were lines 3, 6, and 5; and the metro stations at very high risk were mainly located on metro lines 6, 3, 5, 1, and 2. Based on fieldwork, we suggest raising exits, installing watertight doors, and using early warning strategies to resist metro floods. This study can provide scientific data for decision-makers to reasonably allocate flood prevention resources, which is significant in reducing flood losses and promoting Guangzhou’s sustainable development. Full article
(This article belongs to the Special Issue Remote Sensing in Urban Natural Hazards Monitoring)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Real pictures of flooding in metro stations and tunnels. (<b>a</b>) The tunnel from Shakoulu and Haitansi station of Zhengzhou metro line 5 (18 July 2021); (<b>b</b>) Shenzhoulu station of Guangzhou metro line 21 (30 July 2021); (<b>c</b>) the 28th Street station on the Lexington Avenue line in Manhattan, New York (2 September 2021); (<b>d</b>) Pudding Mill Lane station of London Docklands Light Rail (26 July 2021).</p> Full article ">Figure 2
<p>Terrain of Guangzhou and its metro system.</p> Full article ">Figure 3
<p>Modeling and procedure of flood risk assessment for metro systems.</p> Full article ">Figure 4
<p>Spatial distribution of hazard indicators in Guangzhou. (<b>a</b>) Rainstorm intensity (≥100 mm); (<b>b</b>) rainstorm intensity (≥50 mm); (<b>c</b>) annual average precipitation; (<b>d</b>) waterlogging points; (<b>e</b>) historical flood frequency. The letters are the abbreviations of the districts in Guangzhou; YX: Yuexiu; TH: Tianhe; LW: Liwan; HZ: Haizhu; BY: Baiyun; HP: Huangpu; PY: Panyu; NS: Nansha; HD: Huadu; ZC: Zengceng; CH: Conghua. These abbreviations will not be explained repeatedly in the rest of the paper.</p> Full article ">Figure 5
<p>Spatial distribution of land subsidence (<b>a</b>) and other geological hazards (<b>b</b>) in Guangzhou.</p> Full article ">Figure 6
<p>Spatial distribution of metro station exits (<b>a</b>), station density (<b>b</b>), and land cover (<b>c</b>) and their flood susceptibility (<b>d</b>) in Guangzhou.</p> Full article ">Figure 7
<p>Spatial distribution of exposure indicators in Guangzhou. (<b>a</b>) DEM; (<b>b</b>) slope; (<b>c</b>); river network density; (<b>d</b>) distance to river network; (<b>e</b>) distance to fault.</p> Full article ">Figure 8
<p>Spatial distribution of vulnerability indicators of Guangzhou. (<b>a</b>) Passenger flow; (<b>b</b>) metro line density; (<b>c</b>) proportion of elderly and children; (<b>d</b>) education level.</p> Full article ">Figure 9
<p>Spatial distribution of vulnerability indicators of Guangzhou. (<b>a</b>) Population density; (<b>b</b>) GDP density; (<b>c</b>) road network density; (<b>d</b>) distance to road network.</p> Full article ">Figure 10
<p>Trapezoidal fuzzy number <span class="html-italic">m</span>. When <span class="html-italic">a</span> = <span class="html-italic">b</span>, the trapezoidal fuzzy number becomes a triangular fuzzy number; when <span class="html-italic">a</span> = <span class="html-italic">b</span> and <span class="html-italic">c</span> = <span class="html-italic">d</span>, the trapezoidal fuzzy number becomes an interval number; when <span class="html-italic">a</span> = <span class="html-italic">b</span> = <span class="html-italic">c</span> = <span class="html-italic">d</span>, the trapezoidal fuzzy number becomes a crisp value.</p> Full article ">Figure 11
<p>Regional flood risk level map of Guangzhou and its results validation.</p> Full article ">Figure 12
<p>Flood risk level map of Guangzhou metro system.</p> Full article ">Figure 13
<p>Area distribution (<b>a</b>) and proportion (<b>b</b>) of different flood risk levels in each district’s metro in Guangzhou. The proportion is the ratio of the area of different flood risk levels in each district to the total area of the corresponding level in the Guangzhou metro system.</p> Full article ">Figure 14
<p>Area distribution of different flood risk levels for each metro line in Guangzhou. VH, H, M, L, and VL denote very high, high, medium, low, and very low flood risk levels, respectively; the larger the circle and number and the redder the color, the greater the area of the risk level.</p> Full article ">Figure 15
<p>Flood risk level distribution of stations on each metro line in Guangzhou. VH, H, M, L, and VL denote very high, high, medium, low, and very low flood risk levels, respectively; the larger the circle and number and the redder the color, the greater the number of stations at this risk level.</p> Full article ">Figure 16
<p>Flood risk level maps for the region (<b>a</b>) and metro system (<b>b</b>) using traditional trapezoidal fuzzy AHP.</p> Full article ">Figure 17
<p>Flood prevention measures for the Guangzhou metro system.</p> Full article ">Figure A1
<p>Flood risk levels of 268 stations in the Guangzhou metro system.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-227
Previous Issue
Next Issue
Back to TopTop