[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
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
Special Issues
Mountain Remote Sensing
remotesensing-logo
Submit to Remote Sensing Review for Remote Sensing Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Mountain Remote Sensing

A special issue of Remote Sensing (ISSN 2072-4292).

Deadline for manuscript submissions: closed (31 August 2018) | Viewed by 85850

Special Issue Editors

Dr. Marc Zebisch

E-Mail Website
Guest Editor
EURAC Research – Institute for Earth Observation, Viale Druso 1, 39100 Bolzano, Italy
Interests: Earth observation and environmental monitoring in mountain regions; impact of climate change; monitoring for informed decision making
Dr. Claudia Notarnicola

E-Mail Website
Guest Editor
EURAC Research – Institute for Earth Observation, Viale Druso 1, 39100 Bolzano, Italy
Interests: retrieval of bio-physical parameters from optical and radar data; multi-sensor data fusion; integrated approach for environmental monitoring in mountain areas
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Mountains are amongst the most vulnerable regions in the world. In the last few decades, mountains worldwide have undergone dramatic changes. Melting glaciers and less snow are leading to changes in the water regime. Natural hazards, such as landslides, rockfalls or glacier lake outbursts, are threatening mountain populations. Land-use change and climate change are putting pressure on the last remaining natural ecosystems, as well as on mountain agriculture and forestry.

Monitoring and understanding these changes, their drivers and impacts are essential to support a sustainable management of the changing mountain environment. In addition, it is a demanding and exciting scientific task. Remote sensing is one of the key methodologies for monitoring mountains, which are often data-scarce regions due to their remoteness and the harsh environment.

With this Special Issue, we would like to give an overview on state-of-the-art remote sensing methodologies and applications in mountain regions and on how remote sensing can contribute to an improved understanding of environmental dynamics in mountains. The latest developments in remote sensing, such as the use of dense time-series of high resolution data, combination of sensors (optical and SAR, multi-resolution), as well as the integration of satellite data with in situ networks should be highlighted. Topics can include:

  • Remote sensing of cryosphere and the water cycle in mountains
  • Remote sensing of natural hazards in mountains
  • Remote sensing of vegetation and land-cover dynamics in mountains.
  • Remote sensing methodologies for mountains (e.g., topographic and atmospheric correction, sensor fusion, etc.)
Dr. Marc Zebisch
Dr. Claudia Notarnicola
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Mountains
  • Snow
  • Glaciers
  • Natural Hazards
  • Mountain forest
  • Mountain agriculture
  • Impact of climate changes

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (12 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
18 pages, 11482 KiB  
Article
Glacier Change, Supraglacial Debris Expansion and Glacial Lake Evolution in the Gyirong River Basin, Central Himalayas, between 1988 and 2015
by Sheng Jiang, Yong Nie, Qiao Liu, Jida Wang, Linshan Liu, Javed Hassan, Xiangyang Liu and Xia Xu
Remote Sens. 2018, 10(7), 986; https://doi.org/10.3390/rs10070986 - 21 Jun 2018
Cited by 43 | Viewed by 8183
Abstract
Himalayan glacier changes in the context of global climate change have attracted worldwide attention due to their profound cryo-hydrological ramifications. However, an integrated understanding of the debris-free and debris-covered glacier evolution and its interaction with glacial lake is still lacking. Using one case [...] Read more.
Himalayan glacier changes in the context of global climate change have attracted worldwide attention due to their profound cryo-hydrological ramifications. However, an integrated understanding of the debris-free and debris-covered glacier evolution and its interaction with glacial lake is still lacking. Using one case study in the Gyirong River Basin located in the central Himalayas, this paper applied archival Landsat imagery and an automated mapping method to understand how glaciers and glacial lakes interactively evolved between 1988 and 2015. Our analyses identified 467 glaciers in 1988, containing 435 debris-free and 32 debris-covered glaciers, with a total area of 614.09 ± 36.69 km2. These glaciers decreased by 16.45% in area from 1988 to 2015, with an accelerated retreat rate after 1994. Debris-free glaciers retreated faster than debris-covered glaciers. As a result of glacial downwasting, supraglacial debris coverage expanded upward by 17.79 km2 (24.44%). Concurrent with glacial retreat, glacial lakes increased in both number (+41) and area (+54.11%). Glacier-connected lakes likely accelerated the glacial retreat via thermal energy transmission and contributed to over 15% of the area loss in their connected glaciers. On the other hand, significant glacial retreats led to disconnections from their proglacial lakes, which appeared to stabilize the lake areas. Continuous expansions in the lakes connected with debris-covered glaciers, therefore, need additional attention due to their potential outbursts. In comparison with precipitation variation, temperature increase was the primary driver of such glacier and glacial lake changes. In addition, debris coverage, size, altitude, and connectivity with glacial lakes also affected the degree of glacial changes and resulted in the spatial heterogeneity of glacial wastage across the Gyirong River Basin. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study area and distribution of glaciers and glacial lakes in 2015. The inset map is from ESRI’s world basemap and the background image was acquired on 25 October 2015 from Landsat 8 OLI. Rectangles 1 and 2 are the areas covered in <a href="#remotesensing-10-00986-f002" class="html-fig">Figure 2</a> and <a href="#remotesensing-10-00986-f003" class="html-fig">Figure 3</a>.</p> Full article ">Figure 2
<p>Spatial patterns of glacier changes between 1988 and 2015.</p> Full article ">Figure 3
<p>Glacier area loss (<b>A</b>) and change rate (<b>B</b>) between 1988 and 2015. The glacier extents shown in both panels were mapped from 1988 imagery.</p> Full article ">Figure 4
<p>Area and frequency of glaciers in various size classes: (<b>A</b>) glacier total areas in the studied six years; (<b>B</b>) C-glaciers in 1988 and 2015; (<b>C</b>) D-glaciers in 1988 and 2015. Glacier numbers are given in green.</p> Full article ">Figure 5
<p>Altitudinal distribution of glaciers and their changes between 1988 and 2015: (<b>A</b>) distributions of glacier areas and the net area loss; (<b>B</b>) area distributions of C-glaciers and D-glaciers; (<b>C</b>) area distributions and changes of C-part and D-part glaciers. The shaded area in (<b>C</b>) is only the area increase in D-parts.</p> Full article ">Figure 6
<p>Area distribution of glaciers at various slope classes in 2015: (<b>A</b>) area distributions of C-glaciers and D-glaciers; (<b>B</b>) area distributions of C-part and D-part of D-glaciers. The average slope was calculated based on DEM with a 30 m spatial resolution for individual C-glacier, D-glacier, C-part and D-part glaciers, and then we summed the area for each class.</p> Full article ">Figure 7
<p>Aspect distributions of glaciers in GRB: (<b>A</b>) areas and counts for all glaciers; (<b>B</b>) areas for C-glaciers and D-glaciers in 1988 and 2015.</p> Full article ">Figure 8
<p>Characteristics of altitudinal distributions and size groups of glacial lakes: (<b>A</b>) altitudinal distributions in area and frequency for glacial lakes in 2015; (<b>B</b>) area and frequency distributions of glacial lakes among size classes in 1988 and 2015.</p> Full article ">Figure 9
<p>Evolution and changing status of typical glacial lakes (labeled by green dots) superimposed on background Landsat images: (<b>A</b>) supraglacial lake on Lalaga Glacier (85.563661°E, 28.426213°N); (<b>B</b>) lake connected with C-glacier (85.522556°E, 28.415646°N); (<b>C</b>) lake connected with D-glacier (85.494453°E, 28.508467°N); (<b>D</b>) unconnected glacier-fed lake (85.170447°E, 28.292374°N); (<b>E</b>) no lake at the glacier terminus (85.593496°E, 28.378879°N); (<b>F</b>) newly-formed lake connected with C-glacier; (<b>G</b>) lake unconnected with C-glacier; (<b>H</b>) lake becoming non-glacier-fed after the disappearance of the upstream glacier.</p> Full article ">Figure 10
<p>Air temperature (<b>A</b>) and annual precipitation (<b>B</b>) changes in the central Himalayas observed from the nearest meteorological stations between 1988 and 2015. The blue line represents the linear regression. The red curve is the 5-year moving average.</p> Full article ">
28 pages, 15886 KiB  
Article
Inventory of Glaciers in the Shaksgam Valley of the Chinese Karakoram Mountains, 1970–2014
by Haireti Alifu, Yukiko Hirabayashi, Brian Alan Johnson, Jean-Francois Vuillaume, Akihiko Kondoh and Minoru Urai
Remote Sens. 2018, 10(8), 1166; https://doi.org/10.3390/rs10081166 - 24 Jul 2018
Cited by 9 | Viewed by 8037
Abstract
The Shaksgam Valley, located on the north side of the Karakoram Mountains of western China, is situated in the transition zone between the Indian monsoon system and dry arid climate zones. Previous studies have reported abnormal behaviors of the glaciers in this region [...] Read more.
The Shaksgam Valley, located on the north side of the Karakoram Mountains of western China, is situated in the transition zone between the Indian monsoon system and dry arid climate zones. Previous studies have reported abnormal behaviors of the glaciers in this region compared to the global trend of glacier retreat, so the region is of special interest for glacier-climatological studies. For this purpose, long-term monitoring of glaciers in this region is necessary to obtain a better understanding of the relationships between glacier changes and local climate variations. However, accurate historical and up-to-date glacier inventory data for the region are currently unavailable. For this reason, this study conducted glacier inventories for the years 1970, 1980, 1990, 2000 and 2014 (i.e., a ~10-year interval) using multi-temporal remote sensing imagery. The remote sensing data used included Corona KH-4A/B (1965–1971), Hexagon KH-9 (1980), Landsat Thematic Mapper (TM) (1990/1993), Landsat Enhanced Thematic Mapper Plus (ETM+) (2000/2001), and Landsat Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) (2014/2015) multispectral satellite images, as well as digital elevation models (DEMs) from the Shuttle Radar Topography Mission (SRTM), DEMs generated from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) images (2005–2014), and Advanced Land Observing Satellite (ALOS) World 3D 30 m mesh (AW3D30). In the year 2014, a total of 173 glaciers (including 121 debris-free glaciers) (>0.5 km2), covering an area of 1478 ± 34 km2 (area of debris-free glaciers: 295 ± 7 km2) were mapped. The multi-temporal glacier inventory results indicated that total glacier area change between 1970–2014 was not significant. However, individual glacier changes showed significant variability. Comparisons of the changes in glacier terminus position indicated that 55 (32 debris-covered) glaciers experienced significant advances (~40–1400 m) between 1970–2014, and 74 (32 debris-covered) glaciers experienced significant advances (~40–1400 m) during the most recent period (2000–2014). Notably, small glaciers showed higher sensitivity to climate changes, and the glaciers located in the western part of the study site were exhibiting glacier area expansion compared to other parts of the Shaksgam Valley. Finally, regression analyses indicated that topographic parameters were not the main driver of glacier changes. On the contrary, local climate variability could explain the complex behavior of glaciers in this region. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>Example of overestimated glacier outlines by the First Chinese Glacier Inventory (FCGI). The overall location of this figure is shown in <a href="#sec2dot1-remotesensing-10-01166" class="html-sec">Section 2.1</a>.</p> Full article ">Figure 2
<p>Comparisons of coverage of glacier inventories from previous studies with this study.</p> Full article ">Figure 3
<p>Utilized data coverage in this study (<b>a</b>) and the overall location of Shaksgam Valley (<b>b</b>).</p> Full article ">Figure 4
<p>Schematic flowchart for generation of multi-temporal glacier inventories.</p> Full article ">Figure 5
<p>Examples of manual digitizing the debris-covered glacier in this study. Debris-covered glacier boundary was identified based on distinct supraglacial debris futures (1–5) using Corona (<b>a</b>) and Hexagon images (<b>b</b>).</p> Full article ">Figure 6
<p>Examples of visual comparisons of glacier outlines derived from this study and references such as Soviet Military Topo maps (<b>a</b>–<b>d</b>), Glacier Area Mapping for Discharge from the Asian Mountains (GAMDAM) glacier inventory (<b>e</b>–<b>f</b>), Glaciers_cci (<b>e</b>–<b>f</b>) and Google Earth™ images (<b>g</b>–<b>h</b>).</p> Full article ">Figure 7
<p>Diagram showing the number, area coverage for different size classes (<b>a</b>), aspect of the glaciers (<b>b</b>) and distribution of glaciers in sub-regions (<b>c</b>). In (<b>a</b>), S and A are representing the mean slope and aspect (°) value of each size classes.</p> Full article ">Figure 8
<p>Glacier area change rate during 1970–2014 ((−): decreased in area, (+): increased in area).</p> Full article ">Figure 9
<p>Spatial patterns of glacier are changes during 1970–2014.</p> Full article ">Figure 10
<p>Glacier area change rate during 1970–1980 (<b>a</b>), 1980–1990 (<b>b</b>), 1990–2000 (<b>c</b>) and 2000–2014 (<b>d</b>) periods ((−): decreased in area, (+): increased in area).</p> Full article ">Figure 11
<p>Spatial patterns of glacier area change during 1970–1980 (<b>a</b>), 1980–1990 (<b>b</b>), 1990–2000 (<b>c</b>) and 2000–2014 (<b>d</b>) periods.</p> Full article ">Figure 12
<p>Variations of glaciers terminus position in Eastern (<b>a</b>), Middle (<b>b</b>) and Western (<b>c</b>) sites of study area during the 1970–2014 period (unit: meter).</p> Full article ">Figure 13
<p>Variations of glaciers terminus position during the 1970–1980 (<b>a</b>–<b>c</b>) and 1980–1990 (<b>d</b>–<b>f</b>) periods (unit: meter).</p> Full article ">Figure 14
<p>Variations of glaciers terminus position during the 1990–2000 (<b>a</b>–<b>c</b>) and 2000–2014 (<b>d</b>–<b>f</b>) periods (unit: meter).</p> Full article ">Figure 15
<p>Comparison of glaciers outlines during 1970–2014 in the eastern part (<b>a</b>), middle part (<b>b</b>) and western part of study area (<b>c</b>).</p> Full article ">Figure 16
<p>Comparisons of glacier outlines and glacier area based on this study, Rankl et al. study and Bhambri et al. study.</p> Full article ">Figure 17
<p>Mean annual (<b>a</b>) and mean monthly (<b>b</b>) cloud cover fraction of Shaksgam Valley derived from Global 1-km Cloud Cover observation database [<a href="#B35-remotesensing-10-01166" class="html-bibr">35</a>].</p> Full article ">Figure 18
<p>Mean annual temperature (<b>a</b>–<b>d</b>), standard deviation of mean annual temperature (<b>a’</b>–<b>d’</b>), mean annual precipitation (<b>e</b>–<b>h</b>) and standard deviation of mean annual precipitation (<b>e’</b>–<b>h’</b>) in the Shaksgam Valley during 1979–2015.</p> Full article ">
32 pages, 8295 KiB  
Article
Multi-Criteria Evaluation of Snowpack Simulations in Complex Alpine Terrain Using Satellite and In Situ Observations
by Jesús Revuelto, Grégoire Lecourt, Matthieu Lafaysse, Isabella Zin, Luc Charrois, Vincent Vionnet, Marie Dumont, Antoine Rabatel, Delphine Six, Thomas Condom, Samuel Morin, Alessandra Viani and Pascal Sirguey
Remote Sens. 2018, 10(8), 1171; https://doi.org/10.3390/rs10081171 - 24 Jul 2018
Cited by 24 | Viewed by 5628
Abstract
This work presents an extensive evaluation of the Crocus snowpack model over a rugged and highly glacierized mountain catchment (Arve valley, Western Alps, France) from 1989 to 2015. The simulations were compared and evaluated using in-situ point snow depth measurements, in-situ seasonal and [...] Read more.
This work presents an extensive evaluation of the Crocus snowpack model over a rugged and highly glacierized mountain catchment (Arve valley, Western Alps, France) from 1989 to 2015. The simulations were compared and evaluated using in-situ point snow depth measurements, in-situ seasonal and annual glacier surface mass balance, snow covered area evolution based on optical satellite imagery at 250 m resolution (MODIS sensor), and the annual equilibrium-line altitude of glaciers, derived from satellite images (Landsat, SPOT, and ASTER). The snowpack simulations were obtained using the Crocus snowpack model driven by the same, originally semi-distributed, meteorological forcing (SAFRAN) reanalysis using the native semi-distributed configuration, but also a fully distributed configuration. The semi-distributed approach addresses land surface simulations for discrete topographic classes characterized by elevation range, aspect, and slope. The distributed approach operates on a 250-m grid, enabling inclusion of terrain shadowing effects, based on the same original meteorological dataset. Despite the fact that the two simulations use the same snowpack model, being potentially subjected to same potential deviation from the parametrization of certain physical processes, the results showed that both approaches accurately reproduced the snowpack distribution over the study period. Slightly (although statistically significantly) better results were obtained by using the distributed approach. The evaluation of the snow cover area with MODIS sensor has shown, on average, a reduction of the Root Mean Squared Error (RMSE) from 15.2% with the semi-distributed approach to 12.6% with the distributed one. Similarly, surface glacier mass balance RMSE decreased from 1.475 m of water equivalent (W.E.) for the semi-distributed simulation to 1.375 m W.E. for the distribution. The improvement, observed with a much higher computational time, does not justify the recommendation of this approach for all applications; however, for simulations that require a precise representation of snowpack distribution, the distributed approach is suggested. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Upper Arve catchment study area. The white shaded area shows the extent of the glaciers in 2012 (Gardent et al., 2014). The inner maps show various magnifications of the Alps and the location of the Arve valley within the mountain range. The red points show the position of the five Météo-France stations located in the study area.</p> Full article ">Figure 2
<p>Glacier surface mass balance (SMB) measurement locations for ablation and accumulation areas in the Mer de Glace and Argentière glaciers.</p> Full article ">Figure 3
<p>Schematic representation of the approaches used to account for mountain spatial heterogeneity when simulating snowpack dynamics.</p> Full article ">Figure 4
<p>Observed (black squares) and simulated (red lines) snow depth evolution for the 2007–2008 (upper panel) and 2012–2013 (bottom panel) snow seasons. The elevations of the stations are: Chamonix: 1025 m.a.s.l.; Le Tour: 1470 m.a.s.l.; La Flegere: 1850 m.a.s.l.; Lognan: 1970 m.a.s.l.; and Aiguilles Rouges: 2365 m.a.s.l.</p> Full article ">Figure 5
<p>Spatial distribution of the Snow Water Equivalent (SWE) simulated with distributed (left panels) and semi-distributed (middle panels) approaches for the 28 February 2003 (upper panels) and for the 24 April 2003 (lower panels). Right maps show the UWS MODImLab product for same dates.</p> Full article ">Figure 6
<p>(<b>Left</b>) Temporal evolution of the SCA (2005–2010) based on semi-distributed and distributed simulations and MODIS sensor observations. The vertical bars associated with the MODIS observations show the uncertainty associated with cloud presence for days having ≤20% snow cover. (<b>Right</b>) Scatter plot between observed (<span class="html-italic">X</span> axis) and simulated (<span class="html-italic">Y</span> axis) SCA obtained with semi-distributed and distributed approaches for the whole time period with observations.</p> Full article ">Figure 7
<p>(<b>Left</b>) Observed and simulated SCA evolution for a period of low level snowpack accumulation (2006–2008; upper panel) and a period of high level snowpack accumulation (2011–2013 lower panel). The vertical bars for the MODIS observations show the uncertainty associated with cloud presence for days having ≤20% snow cover. Red and blue shading for the distributed and semi-distributed SCA simulations show the uncertainty associated with various snow depth thresholds for determining whether a pixel was snow covered. The lower limit of the shading represents the SCA evolution for a 0.1 m threshold, the upper limit of the shading represents a 0.2 m snow depth threshold, and the middle line represents a 0.15 m snow depth threshold. (<b>Right</b>) Scatter plot between observed (<span class="html-italic">X</span> axis) and simulated (<span class="html-italic">Y</span> axis) SCA obtained with semi-distributed and distributed approaches for same time period of the left graphs (respectively 2006–2008 and 2011–2013).</p> Full article ">Figure 8
<p>Evolution of the SCA in relation to north and south aspect for the 2006–2008 (<b>upper panel</b>; low level of snowpack accumulation) and 2011–2013 (<b>lower panel</b>; high level of snowpack accumulation) snow seasons. Vertical bars for the MODIS observations show the uncertainty associated with cloud presence for days having ≤20% snow cover. Red and blue shading for the distributed and semi-distributed SCA simulations show the uncertainty associated with various snow depth thresholds for determining whether a pixel was snow covered. The lower limit of the shading represents the SCA evolution for a 0.1 m threshold, the upper limit of the shading represents a 0.2 m snow depth threshold, and the middle line represents a 0.15 m snow depth threshold.</p> Full article ">Figure 9
<p>Jaccard index and ASSD values for low level (2006–2007 and 2007–2008) and high level (2011–2012 and 2012–2013) snow accumulation seasons.</p> Full article ">Figure 10
<p>Temporal evolution of the observed and simulated (semi-distributed and distributed) SMB for the Argentière glacier for the four 300-m elevation bands for the period 1994–2013. The points show the average observation and simulation values for the same measurement locations, and the vertical bars show the standard deviations for those values.</p> Full article ">Figure 11
<p>Temporal evolution of the observed and simulated (semi-distributed and distributed) SMB for the Mer de Glace glacier for the seven 300-m elevations bands for the period 1994–2013. The points show the average observation and simulation values for the same measurement locations, and the vertical bars show the standard deviations for those values.</p> Full article ">Figure 12
<p>Altitudinal dependence of the observed and simulated (semi-distributed and distributed) SMB for two snow seasons (2007–2008: low level snow accumulation; and 2012–2013: high level snow accumulation) at the Mer de Glace glacier.</p> Full article ">Figure 13
<p>Observed and simulated evolution of the equilibrium-line altitude (ELA) for the five glaciers during the study period, based on the same dates as those for the satellite image acquisition.</p> Full article ">
21 pages, 4120 KiB  
Article
Evaluating the Best Spectral Indices for the Detection of Burn Scars at Several Post-Fire Dates in a Mountainous Region of Northwest Yunnan, China
by Davide Fornacca, Guopeng Ren and Wen Xiao
Remote Sens. 2018, 10(8), 1196; https://doi.org/10.3390/rs10081196 - 30 Jul 2018
Cited by 79 | Viewed by 7723
Abstract
Remote mountainous regions are among the Earth’s last remaining wild spots, hosting rare ecosystems and rich biodiversity. Because of access difficulties and low population density, baseline information about natural and human-induced disturbances in these regions is often limited or nonexistent. Landsat time series [...] Read more.
Remote mountainous regions are among the Earth’s last remaining wild spots, hosting rare ecosystems and rich biodiversity. Because of access difficulties and low population density, baseline information about natural and human-induced disturbances in these regions is often limited or nonexistent. Landsat time series offer invaluable opportunities to reconstruct past land cover changes. However, the applicability of this approach strongly depends on the availability of good quality, cloud-free images, acquired at a regular time interval, which in mountainous regions are often difficult to find. The present study analyzed burn scar detection capabilities of 11 widely used spectral indices (SI) at 1 to 5 years after fire events in four dominant vegetation groups in a mountainous region of northwest Yunnan, China. To evaluate their performances, we used M-statistic as a burned-unburned class separability index, and we adapted an existing metric to quantify the SI residual burn signal at post-fire dates compared to the maximum severity recorded soon after the fire. Our results show that Normalized Burn Ratio (NBR) and Normalized Difference Moisture Index (NDMI) are always among the three best performers for the detection of burn scars starting 1 year after fire but not for the immediate post-fire assessment, where the Mid Infrared Burn Index, Burn Area Index, and Tasseled Cap Greenness were superior. Brightness and Wetness peculiar patterns revealed long-term effects of fire in vegetated land, suggesting their potential integration to assist other SI in burned area detection several years after the fire event. However, in general, class separability of most of the SI was poor after one growing season, due to the seasonal rains and the relatively fast regrowth rate of shrubs and grasses, confirming the difficulty of assessment in mountainous ecosystems. Our findings are meaningful for the selection of a suitable SI to integrate in burned area detection workflows, according to vegetation type and time lag between image acquisitions. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Map of northwest Yunnan and location of burn plots. The land cover classes in this map of northwest Yunnan have been aggregated from the European Space Agency Climate Change Initiative (ESA CCI) Landcover product (reference in the core text, <a href="#sec2dot2-remotesensing-10-01196" class="html-sec">Section 2.2</a>), and are represented over a shaded relief layer. The four rivers flowing in parallel are, from West to East: Dulongjiang, Nujiang, Lancangjiang, and Jinshajiang. Detailed information of the 12 burn plots are given in <a href="#remotesensing-10-01196-t001" class="html-table">Table 1</a>.</p> Full article ">Figure 2
<p>Temporal trajectories of burned and unburned vegetation. This conceptual model represents the effect of fire on vegetation as well as severity and recovery patterns. Spectral Indices serve as proxies to represent the ecological condition of the vegetation. Modified and adapted from Key [<a href="#B65-remotesensing-10-01196" class="html-bibr">65</a>] and Chompuchan &amp; Lin [<a href="#B64-remotesensing-10-01196" class="html-bibr">64</a>].</p> Full article ">Figure 3
<p>Burned vs. unburned class distributions for the NDVI at pre-fire, post-fire, and 1, 2, 5 years after fire. An example of burn scar, burn and control sample points, as well as the <span class="html-italic">M</span>-statistic results are illustrated.</p> Full article ">Figure 4
<p>Spectral Indices’ trajectories over time: pre- and post-fire, and 1, 2, 3, 4, and 5 years following the fire. The plots represent the SI mean values for the burned and the reference samples of the whole dataset.</p> Full article ">Figure 5
<p>Temporal trajectories for 11 spectral indices in four major vegetation types in northwest Yunnan.</p> Full article ">
18 pages, 5517 KiB  
Article
Ecosystem Services in a Protected Mountain Range of Portugal: Satellite-Based Products for State and Trend Analysis
by Claudia Carvalho-Santos, António T. Monteiro, Salvador Arenas-Castro, Felix Greifeneder, Bruno Marcos, Ana Paula Portela and João Pradinho Honrado
Remote Sens. 2018, 10(10), 1573; https://doi.org/10.3390/rs10101573 - 1 Oct 2018
Cited by 17 | Viewed by 5732
Abstract
Mountains are facing strong environmental pressures, which may jeopardize the supply of various ecosystem services. For sustainable land management, ecosystem services and their supporting functions should thus be evaluated and monitored. Satellite products have been receiving growing attention for monitoring ecosystem functioning, mainly [...] Read more.
Mountains are facing strong environmental pressures, which may jeopardize the supply of various ecosystem services. For sustainable land management, ecosystem services and their supporting functions should thus be evaluated and monitored. Satellite products have been receiving growing attention for monitoring ecosystem functioning, mainly due to their increasing temporal and spatial resolutions. Here, we aim to illustrate the high potential of satellite products, combined with ancillary in situ and statistical data, to monitor the current state and trend of ecosystem services in the Peneda-Gerês National Park, a protected mountain range in Portugal located in a transition climatic zone (Atlantic to Mediterranean). We focused on three ecosystem services belonging to three broad categories: provisioning (reared animals), regulating (of water flows), and cultural (conservation of an endemic and iconic species). These services were evaluated using a set of different satellite products, namely grassland cover, soil moisture, and ecosystem functional attributes. In situ and statistical data were also used to compute final indicators of ecosystem services. We found a decline in the provision of reared animals since year 2000, although the area of grasslands had remained stable. The regulation of water flows had been maintained, and a strong relationship with interannual precipitation pattern was noted. In the same period, conservation of the focal iconic species might have been affected by interannual fluctuations of suitable habitat areas, with a possible influence of wildfires and precipitation. We conclude that satellite products can efficiently provide information about the current state and trend in the supply of various categories of ecosystem services, especially when combined with in situ or statistical data in robust modeling frameworks. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Location of Peneda-Gerês National Park (PGNP). (<b>A</b>) Elevation in PGNP based on EU-DEM (European Environmental Agency). (<b>B</b>) Portugal and Europe. (<b>C</b>) Northern Portugal northwestern Spain.</p> Full article ">Figure 2
<p>Comparison of volumetric soil moisture (SM) between Sentinel-1 retrievals and in-field measurements in PGNP for three dates where Sentinel-1 passed the region during the summer of 2016 (7th of June, 12th of July, and 21st of September).</p> Full article ">Figure 3
<p>(<b>A</b>) Reared animals service provision demonstrated by the evolution in the number of cattle heads from 1999 to 2009 and the grassland area cover (ha) in 2002 and 2016 in the parishes of PGNP. (<b>B</b>) Grazing livestock density index (GLDI) in the parishes of PGNP, illustrating the level of cattle pressure in the grassland cover area in 2002 and 2016.</p> Full article ">Figure 4
<p>Sentinel-1 soil moisture (%) in PGNP. (<b>A</b>) Spatial distribution of soil moisture in two dates of the month of February 2015 and 2016, respectively. Cells with no data (white) refer to cells that the algorithm does not retrieve, such as forest and bare rock land cover classes. (<b>B</b>) Average monthly Sentinel-1 soil moisture for the years 2015, 2016, and 2017 with average monthly local precipitation (mm) and temperature (°C) from SNIRH (snirh.apambiente.pt).</p> Full article ">Figure 5
<p>Trend in ESA CCI SM [<a href="#B28-remotesensing-10-01573" class="html-bibr">28</a>] and precipitation from E-OBS [<a href="#B35-remotesensing-10-01573" class="html-bibr">35</a>] for the period 2000–2016 in PGNP.</p> Full article ">Figure 6
<p>(<b>A</b>) Distribution of predicted suitable areas (PSA) between 2001 and 2007 by classes for each grid unit (1 km<sup>2</sup>) for <span class="html-italic">Iris boissieri</span> in PGNP. Pictures from <span class="html-italic">Iris boissieri</span> and suitable habitat gently provided by ADERE association (<a href="http://www.adere-pg.pt/?lg=2" target="_blank">http://www.adere-pg.pt/?lg=2</a>). (<b>B</b>) Response curves of the two main contributing predictors: EVImn (enhanced vegetation index annual minimum) and LSTdmxc (cosine of the date of maximum land surface temperature) with PSA. (<b>C</b>) Interannual variation of PSA (black line) and PSA predictions for 2007, year with observed presences (dashed red line), along with auxiliary data related to the total amount of burned area in the study site each year (grey bars).</p> Full article ">
25 pages, 5509 KiB  
Article
Impacts of Climate and Supraglacial Lakes on the Surface Velocity of Baltoro Glacier from 1992 to 2017
by Anna Wendleder, Peter Friedl and Christoph Mayer
Remote Sens. 2018, 10(11), 1681; https://doi.org/10.3390/rs10111681 - 24 Oct 2018
Cited by 20 | Viewed by 6705
Abstract
The Baltoro Glacier is one of the largest glaciers in the Karakoram mountain range. Long-term monitoring of glacier dynamics provides key information on glacier evolution in a changing climate, which is essential for regional water resource and natural hazard management. On large glaciers, [...] Read more.
The Baltoro Glacier is one of the largest glaciers in the Karakoram mountain range. Long-term monitoring of glacier dynamics provides key information on glacier evolution in a changing climate, which is essential for regional water resource and natural hazard management. On large glaciers, detailed field based mass balance is not feasible. Ice dynamic variations quantify changes in mass transport and possibly the influence of environmental parameters on the evolution of the glacier. Although velocity variations of Baltoro Glacier during winter and summer are linked to seasonally enhanced basal sliding, little is known about differences in timing and magnitude of (intra-)seasonal velocity variations and their determining mechanisms. We present time series of annual, seasonal, and intra-seasonal glacier surface velocities by means of intensity offset tracking applied on multi-mission Synthetic Aperture Radar (SAR) data for a period of 25 years from 1992 to 2017. Supraglacial lakes forming on the downstream glacier surface in summer were mapped from 1991 to 2017 based on the Normalized Difference Water Index (NDWI), calculated from multi-spectral Landsat and ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) imagery. Additionally, precipitation data of the Tropical Rainfall Measurement Mission (TRMM) and temperature data of ERA-Interim were used to derive the Standardized Precipitation Index (SPI) and Standardized Temperature Index (STI) from 1998 to 2017. Linking surface velocities to the SPI confirmed a strong correlation between heavy precipitation events in winter and the magnitude and the timing of glacier acceleration in summer. Downstream extensions of summer acceleration that have been found since 2015 may be explained by additional water draining from an increased number of supraglacial lakes through crevasses that have been formed in consequence of higher initial velocities, evoked by strong winter precipitation. The warmer melt seasons observed in the years 2015 to 2017 additionally affects the formation of a supraglacial lake, so stronger summer acceleration events in recent years may be indirectly related to global warming. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Overview of Baltoro Glacier, Pakistan, with the glacier boundaries (blue) and the glacier centerline (red). The background image is a Landsat-7 scene acquired on 12 July 2010, which has been also used for derivation of the glacier boundaries based on the Normalized Difference Snow index (NDSI) and manual editing.</p> Full article ">Figure 2
<p>Annual surface velocities (m a<sup>−1</sup>) along the glacier centerline.</p> Full article ">Figure 3
<p>Averaged summer surface velocities (m a<sup>−1</sup>) along the glacier’s centerline.</p> Full article ">Figure 4
<p>Averaged winter surface velocities (m a<sup>−1</sup>) along the glacier’s centerline.</p> Full article ">Figure 5
<p>Heatmap of summer surface velocities (m a<sup>−1</sup>) from May to September along the glacier centerline for 1999–2017.</p> Full article ">Figure 6
<p>Heatmap of winter surface velocities (m a<sup>−1</sup>) from October to April along the glacier centerline.</p> Full article ">Figure 7
<p>SPI from 1998 to 2017 for the monthly precipitation intensity, color-coded for the seasonal distribution: January and February (JF) in dark blue, March to May (MAM) in light blue, June to September (JJAS) in red, and October to December (OND) in orange.</p> Full article ">Figure 8
<p>STI from 1998 to 2017 for the monthly temperature anomaly, color-coded for the seasonal distribution: January and February (JF) in dark blue, March to May (MAM) in light blue, June to September (JJAS) in red, and October to December (OND) in orange.</p> Full article ">Figure A1
<p>Overview of the Baltoro Glacier, Pakistan, with the 121 areas on stable, non-moving, and non-glaciated ground that were used for the uncertainty analysis of the surface velocities. The background image is a Landsat-7 scene acquired on 12 July 2010, which is also used for the digitalization of the glacier boundaries.</p> Full article ">Figure A2
<p>Averaged velocity field of all summer and winter surface velocities (m a<sup>−1</sup>) for the period 1992 to 2017.</p> Full article ">Figure A3
<p>Averaged velocity field of all summer surface velocities (m a<sup>−1</sup>) for the period 2015 to 2017.</p> Full article ">
26 pages, 7265 KiB  
Article
Relationship between Spatiotemporal Variations of Climate, Snow Cover and Plant Phenology over the Alps—An Earth Observation-Based Analysis
by Sarah Asam, Mattia Callegari, Michael Matiu, Giuseppe Fiore, Ludovica De Gregorio, Alexander Jacob, Annette Menzel, Marc Zebisch and Claudia Notarnicola
Remote Sens. 2018, 10(11), 1757; https://doi.org/10.3390/rs10111757 - 7 Nov 2018
Cited by 44 | Viewed by 6490
Abstract
Alpine ecosystems are particularly sensitive to climate change, and therefore it is of significant interest to understand the relationships between phenology and its seasonal drivers in mountain areas. However, no alpine-wide assessment on the relationship between land surface phenology (LSP) patterns and its [...] Read more.
Alpine ecosystems are particularly sensitive to climate change, and therefore it is of significant interest to understand the relationships between phenology and its seasonal drivers in mountain areas. However, no alpine-wide assessment on the relationship between land surface phenology (LSP) patterns and its climatic drivers including snow exists. Here, an assessment of the influence of snow cover variations on vegetation phenology is presented, which is based on a 17-year time-series of MODIS data. From this data snow cover duration (SCD) and phenology metrics based on the Normalized Difference Vegetation Index (NDVI) have been extracted at 250 m resolution for the entire European Alps. The combined influence of additional climate drivers on phenology are shown on a regional scale for the Italian province of South Tyrol using reanalyzed climate data. The relationship between vegetation and snow metrics strongly depended on altitude. Temporal trends towards an earlier onset of vegetation growth, increasing monthly mean NDVI in spring and late summer, as well as shorter SCD were observed, but they were mostly non-significant and the magnitude of these tendencies differed by altitude. Significant negative correlations between monthly mean NDVI and SCD were observed for 15–55% of all vegetated pixels, especially from December to April and in altitudes from 1000–2000 m. On the regional scale of South Tyrol, the seasonality of NDVI and SCD achieved the highest share of correlating pixels above 1500 m, while at lower elevations mean temperature correlated best. Examining the combined effect of climate variables, for average altitude and exposition, SCD had the highest effect on NDVI, followed by mean temperature and radiation. The presented analysis allows to assess the spatiotemporal patterns of earth-observation based snow and vegetation metrics over the Alps, as well as to understand the relative importance of snow as phenological driver with respect to other climate variables. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of the study area showing the elevation with the boundary of the Alps as defined by the Alpine Convention (green), state borders (white), and the border of the province of Bolzano (South Tyrol, red).</p> Full article ">Figure 2
<p>Selected land cover classes of the COoRdination of INformation on the Environment (CORINE) Land Cover product over the Alps.</p> Full article ">Figure 3
<p>Schematic representation of the different pixel-wise correlation approaches.</p> Full article ">Figure 4
<p>Schematic representation of the different single parameter cross-correlations. Abbreviations: NDVI, (normalized difference vegetation index), pre (precipitation), rad (radiation), SCD (snow cover duration), and tmean (mean temperature).</p> Full article ">Figure 5
<p>Start of season (SOS) metric for 2016 over the Alps (<b>left</b>), South Tyrol (<b>upper right</b>) and the Ahrn Valley (<b>lower right</b>).</p> Full article ">Figure 6
<p>Yearly alpine-wide median SOS of the years 2001–2017 for the ten analyzed CORINE land covers. The amount of observations available per land cover class, year and altitude range is indicated by the transparency level of each dot, see “log(Count)”.</p> Full article ">Figure 7
<p>Monthly alpine-wide median NDVI of the years 2000–2017 in different altitude ranges. The formula, correlation coefficient and accuracy level are indicated for the median NDVI averaged over all altitudes.</p> Full article ">Figure 8
<p>Spatial patterns of yearly SCD for the years 2010 (<b>left</b>) and 2016 (<b>right</b>) over the Alps.</p> Full article ">Figure 9
<p>Mean SCD of different months and periods averaged over all altitudinal ranges over the years 2000–2017.</p> Full article ">Figure 10
<p>Yearly median SCD of the Alps in ten different altitudes over the years 2000–2017.</p> Full article ">Figure 11
<p>Spatial representation of significantly positively (red) and negatively (blue) correlated SCD and NDVI in February (<b>left</b>) and September (<b>right</b>).</p> Full article ">Figure 12
<p>Pearson correlation coefficients for the different negative correlations of mean NDVI and SCD of the respective same month over different altitude ranges.</p> Full article ">Figure 13
<p>Percentages of negatively and positively correlating pixels of the December (violet), January (orange) and February (red), longer winter (green) and shorter winter (blue) SCD with the following monthly mean NDVIs.</p> Full article ">Figure 14
<p>Tabular overview on the relative amount of significantly negatively (<b>top</b>) and positively (<b>bottom</b>) correlating pixels of winter SCD (December–February) with all other months’ mean NDVI.</p> Full article ">Figure 14 Cont.
<p>Tabular overview on the relative amount of significantly negatively (<b>top</b>) and positively (<b>bottom</b>) correlating pixels of winter SCD (December–February) with all other months’ mean NDVI.</p> Full article ">Figure 15
<p>Common seasonality of NDVI with climate. For each 100 m altitude class are shown the percentages of climate variables that have the highest correlation with NDVI in South Tyrol (percentages of grid cells in each altitude class). The correlations have been assessed with possible time lags (see also <a href="#app1-remotesensing-10-01757" class="html-app">Figure S8</a>), and here the best correlating variable of all time lags is shown.</p> Full article ">Figure 16
<p>Influence of climate variability measured in the respective unit days [d], degree Celsius [C°], and watt per square meter [W/m<sup>2</sup>] on NDVI at selected dates throughout the year and depending on altitude. Shown are effects (slopes/coefficient of the regression models) of deseasonalized climate variables on deseasonalized NDVI at three altitudes (700, 1500, and 2300 m, which correspond approximately to the 5, 50 and 95% quantiles) and heat load index of 0.75 (approximately the sample average), holding other variables constant. If lines for 700 m and 2300 m are missing, the interaction with altitude was not significant. In empty panels the coefficient of the climate variable was non-significant or, in the case of Winter SCD in January, not included in the model. Shaded areas denote 95% confidence intervals. For more details see <a href="#app1-remotesensing-10-01757" class="html-app">Figures S7 and S8</a>.</p> Full article ">
26 pages, 21464 KiB  
Article
Performance Assessment of TanDEM-X DEM for Mountain Glacier Elevation Change Detection
by Julian Podgórski, Christophe Kinnard, Michał Pętlicki and Roberto Urrutia
Remote Sens. 2019, 11(2), 187; https://doi.org/10.3390/rs11020187 - 18 Jan 2019
Cited by 30 | Viewed by 6591
Abstract
TanDEM-X digital elevation model (DEM) is a global DEM released by the German Aerospace Center (DLR) at outstanding resolution of 12 m. However, the procedure for its creation involves the combination of several DEMs from acquisitions spread between 2011 and 2014, which casts [...] Read more.
TanDEM-X digital elevation model (DEM) is a global DEM released by the German Aerospace Center (DLR) at outstanding resolution of 12 m. However, the procedure for its creation involves the combination of several DEMs from acquisitions spread between 2011 and 2014, which casts doubt on its value for precise glaciological change detection studies. In this work we present TanDEM-X DEM as a high-quality product ready for use in glaciological studies. We compare it to Aerial Laser Scanning (ALS)-based dataset from April 2013 (1 m), used as the ground-truth reference, and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) V003 DEM and SRTM v3 DEM (both 30 m), serving as representations of past glacier states. We use a method of sub-pixel coregistration of DEMs by Nuth and Kääb (2011) to determine the geometric accuracy of the products. In addition, we propose a slope-aspect heatmap-based workflow to remove the errors resulting from radar shadowing over steep terrain. Elevation difference maps obtained by subtraction of DEMs are analyzed to obtain accuracy assessments and glacier mass balance reconstructions. The vertical accuracy (± standard deviation) of TanDEM-X DEM over non-glacierized area is very good at 0.02 ± 3.48 m. Nevertheless, steep areas introduce large errors and their filtering is required for reliable results. The 30 m version of TanDEM-X DEM performs worse than the finer product, but its accuracy, −0.08 ± 7.57 m, is better than that of SRTM and ASTER. The ASTER DEM contains errors, possibly resulting from imperfect DEM creation from stereopairs over uniform ice surface. Universidad Glacier has been losing mass at a rate of −0.44 ± 0.08 m of water equivalent per year between 2000 and 2013. This value is in general agreement with previously reported mass balance estimated with the glaciological method for 2012–2014. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Figure 1

Figure 1
<p>Spatial and temporal extent of the study. (<b>A</b>)—study site location; (<b>B</b>)—temporal distribution of SAR acquisitions used to build TanDEM. The acquisitions dates (dotted lines) were extracted from the DEM metadata. The solid line indicates the acquisition date of the ALS DEM used as reference. (<b>C</b>)—Universidad Glacier topography, with hill-shaded ALS DEM used as basemap. The extent of GA is outlined in blue; (<b>D</b>)—extent of all glacierized areas within the study area, with hill-shaded TanDEM DEM 12 m used as basemap. Areas falling outside the green outline and within the ALS DEM extent were considered stable area (SA).</p> Full article ">Figure 2
<p>Artifacts of ASTER DEM on Universidad Glacier: (<b>A</b>): ASTER true-color composition (9 April 2003); (<b>B</b>): ASTER DEM hill-shade; (<b>C</b>): SRTM DEM hill-shade.</p> Full article ">Figure 3
<p>Results of data analysis and corrections of TanDEM12 relative to the ALS on SA (<b>A</b>–<b>E</b>) and applied to GA (<b>F</b>–<b>H</b>). (<b>A</b>,<b>B</b>): distribution of <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z<math display="inline"> <semantics> <mrow> <mo form="prefix">tan</mo> <mo>(</mo> <mi>α</mi> <mo>)</mo> </mrow> </semantics> </math> vs. aspect <math display="inline"> <semantics> <mi mathvariant="sans-serif">Ψ</mi> </semantics> </math> before and after GC respectively; (<b>C</b>): <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z histograms before and after GC; (<b>D</b>): variation of <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z with slope of the surface. (<b>E</b>–<b>G</b>): heatmaps of median <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z in slope-aspect (<math display="inline"> <semantics> <mrow> <mi>α</mi> <mo>−</mo> <mi mathvariant="sans-serif">Ψ</mi> </mrow> </semantics> </math>) domain binned in 1<math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math> <math display="inline"> <semantics> <mi>α</mi> </semantics> </math> and 4<math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math> <math display="inline"> <semantics> <mi mathvariant="sans-serif">Ψ</mi> </semantics> </math> bins. (<b>E</b>): SA; (<b>F</b>): GA before slope/aspect correction; (<b>G</b>): GA after SAHC; The extreme values of the color scale correspond to maximum and minimum median <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z across all three heatmaps. (<b>H</b>): 2D histogram of the slope and aspect distribution across the GA.</p> Full article ">Figure 4
<p>Results of data analysis and corrections of TanDEM30 relative to the ALS on SA (<b>A</b>–<b>E</b>) and applied to GA (<b>F</b>–<b>H</b>). (<b>A</b>,<b>B</b>): distribution of <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z<math display="inline"> <semantics> <mrow> <mo form="prefix">tan</mo> <mo>(</mo> <mi>α</mi> <mo>)</mo> </mrow> </semantics> </math> vs. aspect <math display="inline"> <semantics> <mi mathvariant="sans-serif">Ψ</mi> </semantics> </math> before and after GC respectively; (<b>C</b>): <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z histograms before and after GC; (<b>D</b>): variation of <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z with slope of the surface. (<b>E</b>–<b>G</b>): heatmaps of median <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z in slope-aspect (<math display="inline"> <semantics> <mrow> <mi>α</mi> <mo>−</mo> <mi mathvariant="sans-serif">Ψ</mi> </mrow> </semantics> </math>) domain binned in 1<math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math> <math display="inline"> <semantics> <mi>α</mi> </semantics> </math> and 4<math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math> <math display="inline"> <semantics> <mi mathvariant="sans-serif">Ψ</mi> </semantics> </math> bins. (<b>E</b>): SA; (<b>F</b>): GA before slope/aspect correction; (<b>G</b>): GA after SAHC; The extreme values of the color scale correspond to maximum and minimum median <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z across all three heatmaps. (<b>H</b>): 2D histogram of the slope and aspect distribution across the GA.</p> Full article ">Figure 5
<p>Results of DEM subtraction with semivariograms and histograms of differences. (<b>A</b>): ALS-TanDEM 12 m; (<b>B</b>): ALS-TanDEM 30 m. Blue dots represent empirical semivariograms, while red lines—theoretical ones. Green vertical lines indicate distance thresholds of the best fit of the model to the data. Black line on the map delimits the GA.</p> Full article ">Figure 6
<p>Results of DEM subtraction with semivariograms and histograms of differences. (<b>A</b>): ALS-SRTM; (<b>B</b>): TanDEM-SRTM. Resolution of both maps is 12 m. Blue dots represent empirical semivariograms, while red lines—synthetic ones. Green vertical line indicates distance threshold of the best fit of the model to the data. Black line on the map delimits the GA, green line corresponds to transect line used for creation of <a href="#remotesensing-11-00187-f008" class="html-fig">Figure 8</a>.</p> Full article ">Figure 7
<p>Results of DEM subtraction with semivariograms and histograms of differences. (<b>A</b>): ALS-ASTER; (<b>B</b>): TanDEM-ASTER. Resolution of both maps is 12 m. Blue dots represent empirical semivariograms, while red lines—synthetic ones. Green vertical line indicates distance threshold of the best fit of the model to the data. Black line on the map delimits the GA.</p> Full article ">Figure 8
<p>Cross-sectional profile of the Universidad Glacier surface elevation in 2000 and 2013 (see profile location in <a href="#remotesensing-11-00187-f006" class="html-fig">Figure 6</a>).</p> Full article ">Figure 9
<p>Vertical profile of measured ice elevation change on Universidad Glacier. All <math display="inline"> <semantics> <mo>Δ</mo> </semantics> </math>Z measurements were averaged per 100 m elevation bin. Only GA pixels were used.</p> Full article ">Figure 10
<p>Polar plot of slope/aspect pairs usable for a single InSAR acquisition (single iDEM) with incidence angle of 31.32<math display="inline"> <semantics> <msup> <mrow/> <mo>∘</mo> </msup> </semantics> </math> over our study area. Red points indicate pixels of ALS, blue circles—slope values corresponding to 90%, 99% and 99.9% of all pixels and shaded area—range of usable slopes obtained with the method of Eineder [<a href="#B42-remotesensing-11-00187" class="html-bibr">42</a>].</p> Full article ">
20 pages, 1193 KiB  
Article
Natural Forest Mapping in the Andes (Peru): A Comparison of the Performance of Machine-Learning Algorithms
by Luis Alberto Vega Isuhuaylas, Yasumasa Hirata, Lenin Cruyff Ventura Santos and Noemi Serrudo Torobeo
Remote Sens. 2018, 10(5), 782; https://doi.org/10.3390/rs10050782 - 18 May 2018
Cited by 40 | Viewed by 7320
Abstract
The Andes mountain forests are sparse relict populations of tree species that grow in association with local native shrubland species. The identification of forest conditions for conservation in areas such as these is based on remote sensing techniques and classification methods. However, the [...] Read more.
The Andes mountain forests are sparse relict populations of tree species that grow in association with local native shrubland species. The identification of forest conditions for conservation in areas such as these is based on remote sensing techniques and classification methods. However, the classification of Andes mountain forests is difficult because of noise in the reflectance data within land cover classes. The noise is the result of variations in terrain illumination resulting from complex topography and the mixture of different land cover types occurring at the sub-pixel level. Considering these issues, the selection of an optimum classification method to obtain accurate results is very important to support conservation activities. We carried out comparative non-parametric statistical analyses on the performance of several classifiers produced by three supervised machine-learning algorithms: Random Forest (RF), Support Vector Machine (SVM), and k-Nearest Neighbor (kNN). The SVM and RF methods were not significantly different in their ability to separate Andes mountain forest and shrubland land cover classes, and their best classifiers showed a significantly better classification accuracy (AUC values 0.81 and 0.79 respectively) than the one produced by the kNN method (AUC value 0.75) because the latter was more sensitive to noisy training data. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Study area in Peru. The red box outlines the image area.</p> Full article ">Figure 2
<p>Preprocessing flowchart.</p> Full article ">Figure 3
<p>Critical Difference (CD) diagram for the Nemenyi test showing the results of the statistical comparison of all models against each other by mean ranks based on AUC values (higher ranks, such as 5.9 for SVM:All, correspond to higher values of AUC). Classifiers that are not connected by a bold line of length equal to CD have significantly different mean ranks (Confidence level of 95%).</p> Full article ">Figure 4
<p>Critical Difference (CD) diagram for the Nemenyi test showing the results of the statistical comparison of all models against each other by mean ranks based on Cohen’s Kappa values (higher ranks, such as 4.9 for SVM:All, correspond to higher values of Cohen’s Kappa). Classifiers that are not connected by a bold line of length equal to CD have significantly different mean ranks (Confidence level of 95%).</p> Full article ">
16 pages, 8790 KiB  
Article
Centimetric Accuracy in Snow Depth Using Unmanned Aerial System Photogrammetry and a MultiStation
by Francesco Avanzi, Alberto Bianchi, Alberto Cina, Carlo De Michele, Paolo Maschio, Diana Pagliari, Daniele Passoni, Livio Pinto, Marco Piras and Lorenzo Rossi
Remote Sens. 2018, 10(5), 765; https://doi.org/10.3390/rs10050765 - 16 May 2018
Cited by 49 | Viewed by 5402
Abstract
Performing two independent surveys in 2016 and 2017 over a flat sample plot (6700 m 2 ), we compare snow-depth measurements from Unmanned-Aerial-System (UAS) photogrammetry and from a new high-resolution laser-scanning device (MultiStation) with manual probing, the standard technique used by operational services [...] Read more.
Performing two independent surveys in 2016 and 2017 over a flat sample plot (6700 m 2 ), we compare snow-depth measurements from Unmanned-Aerial-System (UAS) photogrammetry and from a new high-resolution laser-scanning device (MultiStation) with manual probing, the standard technique used by operational services around the world. While previous comparisons already used laser scanners, we tested for the first time a MultiStation, which has a different measurement principle and is thus capable of millimetric accuracy. Both remote-sensing techniques measured point clouds with centimetric resolution, while we manually collected a relatively dense amount of manual data (135 pt in 2016 and 115 pt in 2017). UAS photogrammetry and the MultiStation showed repeatable, centimetric agreement in measuring the spatial distribution of seasonal, dense snowpack under optimal illumination and topographic conditions (maximum RMSE of 0.036 m between point clouds on snow). A large fraction of this difference could be due to simultaneous snowmelt, as the RMSE between UAS photogrammetry and the MultiStation on bare soil is equal to 0.02 m. The RMSE between UAS data and manual probing is in the order of 0.20–0.30 m, but decreases to 0.06–0.17 m when areas of potential outliers like vegetation or river beds are excluded. Compact and portable remote-sensing devices like UASs or a MultiStation can thus be successfully deployed during operational manual snow courses to capture spatial snapshots of snow-depth distribution with a repeatable, vertical centimetric accuracy. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Topography of the surveyed area during summer 2016 ((<b>a</b>), top-left), winter 2016 ((<b>b</b>), top-right) and winter 2017 ((<b>c</b>), bottom-right). The boundaries of the study area are in black. Red dots denote the station points of the MultiStation, whereas the blue triangle represents the point used as angular reference. No bare-soil survey was performed during summer 2017.</p> Full article ">Figure 2
<p>Digital Surface Models from the UAS surveys during summer 2016 ((<b>a</b>), top-left), winter 2016 ((<b>b</b>), top-right) and winter 2017 ((<b>c</b>), bottom-right). The color map represents surface height (ASL). Contour lines (grey) are reported with an equidistant interval of 2 m. No bare-soil survey was performed during summer 2017.</p> Full article ">Figure 3
<p>Height differences between the MultiStation (MS) dataset and the photogrammetric products for all surveys. C1 compares the MultiStation and the photogrammetric point clouds, whereas C2 compares the MultiStation point cloud with the DSM from UAS data.</p> Full article ">Figure 4
<p>Histograms of the residuals between the MultiStation scan and the DSM from UAS data (C2) for each of the three surveys (<b>a</b>). Scatter plot of UAS-based vs. MS-based heights (ASL) for both winter and summer cases (<b>b</b>).</p> Full article ">Figure 5
<p>Evaluation of UAS data using manual probing. (<b>a</b>,<b>b</b>): Scatter plots between UAS-based and manual measurements of HS for 2016 and 2017, respectively (regression line in red). (<b>c</b>,<b>d</b>): Spatial distribution of the differences between UAS-based and manual measurements of snow depth. The color legend represents the differences between UAS and manual probing. (<b>e</b>): Histograms of the differences between UAS-based and manual measurements of snow depth for both surveys.</p> Full article ">
16 pages, 7361 KiB  
Article
Fusion of NASA Airborne Snow Observatory (ASO) Lidar Time Series over Mountain Forest Landscapes
by António Ferraz, Sassan Saatchi, Kat J. Bormann and Thomas H. Painter
Remote Sens. 2018, 10(2), 164; https://doi.org/10.3390/rs10020164 - 24 Jan 2018
Cited by 13 | Viewed by 5660
Abstract
Mountain ecosystems are among the most fragile environments on Earth. The availability of timely updated information on forest 3D structure would improve our understanding of the dynamic and impact of recent disturbance and regeneration events including fire, insect damage, and drought. Airborne lidar [...] Read more.
Mountain ecosystems are among the most fragile environments on Earth. The availability of timely updated information on forest 3D structure would improve our understanding of the dynamic and impact of recent disturbance and regeneration events including fire, insect damage, and drought. Airborne lidar is a critical tool for monitoring forest change at high resolution but it has been little used for this purpose due to the scarcity of long-term time-series of measurements over a common region. Here, we investigate the reliability of on-going, multi-year lidar observations from the NASA-JPL Airborne Snow Observatory (ASO) to characterize forest 3D structure at a fine spatial scale. In this study, weekly ASO measurements collected at ~1 pt/m2, primarily acquired to quantify snow volume and dynamics, are coherently merged to produce high-resolution point clouds ( ~ 12 pt/m2) that better describe forest structure. The merging methodology addresses the spatial bias in multi-temporal data due to uncertainties in platform trajectory and motion by collecting tie objects from isolated tree crown apexes in the lidar data. The tie objects locations are assigned to the centroid of multi-temporal lidar points to fuse and optimize the location of multiple measurements without the need for ancillary data or GPS control points. We apply the methodology to ASO lidar acquisitions over the Tuolumne River Basin in the Sierra Nevada, California, during the 2014 snow monitoring campaign and provide assessment of the fidelity of the fused point clouds for forest mountain ecosystem studies. The availability of ASO measurements that currently span 2013–2017 enable annual forest monitoring of important vegetated ecosystems that currently face ecological threads of great significance such as the Sierra Nevada (California) and Olympic National Forest (Washington). Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Airborne Snow Observatory (ASO) coverage sites (black dots) across the US overlapped to a ASTER SRTM hillshade map. In (<b>b</b>) we focus on the Tuolumne River basin (Sierra Nevada, Central California) study site (the black polygon), which is overlapping a 10 m resolution DEM from the USGS National Elevation Dataset (NED) provided by Google Earth (Tuolumne River basin, 37°52′32.16″N, 119°20′08.85″W, <a href="http://www.earth.google.com" target="_blank">http://www.earth.google.com</a> (26 May 2014)). Well known locations such as Mono Lake, Mount Dana, and Excelsior Mountain are provided for better location. The study site (7 km × 4 km) is represented by the blue rectangle.</p> Full article ">Figure 2
<p>Method workflow to calculate the normalized high-density point cloud from multi-temporal low-resolution data. GDT stand for Ground Delaunay Triangulation. Method steps are represented by the gray rectangles.</p> Full article ">Figure 3
<p>Schematic of the lidar registration methodology showing (<b>a</b>) two isolated trees selected as tie objects (green trees referred as T1 and T2) and the highest lidar points (called vertices) from three different lidar acquisitions (referred as A1, A2, and A3) that define the location of the corresponding trees (orange, cyan and blue trees), which are offset from the apex of the tie object, (i.e., the green tree in the foreground). In (<b>b</b>), we represent the location of the original vertices and corresponding centroids (red cross) that are used to calculated the adjusted (or registered) vertices. The colored large semi-transparent circles follow the colors in (<b>a</b>) and represent the trees from an aerial view. The diameter of the circles is figurative and has no physical meaning. In (<b>c</b>), we show the coordinates of the adjusted vertices and corresponding centroids that are reduced to a local referential (with origin in (0,0)). The coordinates of the centroid give us an estimation of the magnitude and direction of the misalignment for each lidar acquisition (please refer to <a href="#sec3dot2-remotesensing-10-00164" class="html-sec">Section 3.2</a> for details regarding (<b>c</b>).</p> Full article ">Figure 4
<p>Tuolumne River basin (<b>a</b>) study site (7 km × 4 km) that corresponds to the blue polygon in <a href="#remotesensing-10-00164-f001" class="html-fig">Figure 1</a>b and the 20 tie objects trees (T1, T2, …, T20) selected for point cloud registration purposes represented by the colored dots and, (<b>b</b>) focus on the orange rectangle in (<b>a</b>) to show three tie object examples over a canopy height model. The colorbar in panel (<b>b</b>) corresponds to the vegetation height.</p> Full article ">Figure 5
<p>Point cloud composites around tie object T1 (<a href="#remotesensing-10-00164-f004" class="html-fig">Figure 4</a>a) (<b>a</b>) before and (<b>b</b>) after the registration of the individual datasets. Colors correspond to acquisition dates. The x and y axis correspond to the UTM coordinates and ellipsoidal altitude, respectively.</p> Full article ">Figure 6
<p>Vertices grouped by acquisition date (small dots) and corresponding centroids (large dots) before (<b>a</b>,<b>b</b>) and after (<b>c</b>,<b>d</b>) registering the point clouds in the horizontal (<b>a</b>,<b>c</b>) and vertical (<b>b</b>,<b>d</b>) dimensions. The coordinates of the vertices represent the distance and direction with respect to the corresponding tie object in planimetry and altimetry (refer to <a href="#remotesensing-10-00164-f003" class="html-fig">Figure 3</a>c). The centroids correspond to the average distance values for each lidar acquisition and are indicators of their systematic bias. Due to the fact that the bias has been removed by our co-registration method, the centroids of the vertices in (<b>c</b>,<b>d</b>) are located over the origin of the corresponding reference (0,0).</p> Full article ">Figure 7
<p>Point cloud calculated (100 m × 100 m) using (<b>a</b>) lower-density single acquisition and (<b>b</b>) a high-density multi-temporal point cloud after co-registration. The canopy height models (1 m resolution, 350 m × 250 m) have been calculated using (<b>c</b>) one and (<b>d</b>) 12 lidar acquisitions.</p> Full article ">Figure 8
<p>Mean canopy height (MCH) calculated using a canopy height model (CHM) derived from increasing lidar acquisitions over area of 1000 m × 1000 m, showing (<b>a</b>) differences over the entire area, (<b>b</b>) when divided into four subsamples (at 500 m × 500 m), and (<b>c</b>) when divided further in 16 subsamples (at 250 × 250 m). The graphs shown are computed by averaging the values calculated by shuffling the order of lidar acquisitions (1–12) five times to remove the impact of eventual bias (e.g., tree growth).</p> Full article ">
4243 KiB  
Article
Performance of Three MODIS Fire Products (MCD45A1, MCD64A1, MCD14ML), and ESA Fire_CCI in a Mountainous Area of Northwest Yunnan, China, Characterized by Frequent Small Fires
by Davide Fornacca, Guopeng Ren and Wen Xiao
Remote Sens. 2017, 9(11), 1131; https://doi.org/10.3390/rs9111131 - 6 Nov 2017
Cited by 91 | Viewed by 9360
Abstract
An increasing number of end-users looking for ground data about fire activity in regions where accurate official datasets are not available adopt a free-of-charge global burned area (BA) and active fire (AF) products for applications at the local scale. One of the pressing [...] Read more.
An increasing number of end-users looking for ground data about fire activity in regions where accurate official datasets are not available adopt a free-of-charge global burned area (BA) and active fire (AF) products for applications at the local scale. One of the pressing requirements from the user community is an improved ability to detect small fires (less than 50 ha), whose impact on terrestrial environments is empirically known but poorly quantified, and is often excluded from global earth system models. The newest generation of BA algorithms combines the capabilities of both the BA and AF detection approaches, resulting in a general improvement of detection compared to their predecessors. Accuracy assessments of these products have been done in several ecosystems; but more complex ones, such as regions that are characterized by frequent small fires and steep terrain has never been assessed. This study contributes to the understanding of the performance of global BA and AF products with a first assessment of four selected datasets: MODIS-based MCD45A1; MCD64A1; MCD14ML; and, ESA’s Fire_CCI in a mountainous region of northwest Yunnan; P.R. China. Due to the medium to coarse resolution of the tested products and the reduced sizes of fires (often smaller than 50 ha) we used a polygon intersection assessment method where the number and locations of fire events extracted from each dataset were compared against a reference dataset that was compiled using Landsat scenes. The results for the two sample years (2006 and 2009) show that the older, non-hybrid products MCD45A1 and, MCD14ML were the best performers with Sørensen index (F1 score) reaching 0.42 and 0.26 in 2006, and 0.24 and 0.24 in 2009, respectively, while producer’s accuracies (PA) were 30% and 43% in 2006, and 16% and 47% in 2009, respectively. All of the four tested products obtained higher probabilities of detection when smaller fires were excluded from the assessment, with PAs for fires bigger than 50 ha being equal to 53% and 61% in 2006, 41% and 66% in 2009 for MCD45A1 and MCD14ML, respectively. Due to the technical limitations of the satellites’ sensors, a relatively low performance of the four products was expected. Surprisingly, the new hybrid algorithms produced worse results than the former two. Fires smaller than 50 ha were poorly detected by the products except for the only AF product. These findings are significant for the future design of improved algorithms aiming for increased detection of small fires in a greater diversity of ecosystems. Full article
(This article belongs to the Special Issue Mountain Remote Sensing)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Location of the study area. The most restrictive boundaries of northwest Yunnan and a broader definition of the region, as well as the Landsat coverage selected for the accuracy assessment are represented. The large map shows the main landcover classes according to MODIS MCD12Q1 product for 2011 over a shaded relief. The four major rivers flowing across the region in a parallel manner on the northern section, from West to East: the Dulongjiang (tributary of the Irrawaddy), the Nujiang (upper Salween), the Lancangjiang (upper Mekong), and the Jinshajiang (upper Yangtze). (MCD12Q1 reference: (<a href="https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1" target="_blank">https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1</a>, last accessed on 25 September 2017).</p> Full article ">Figure 2
<p>Example of polygon intersection assessment: (<b>1</b>) Burn scar visible on Landsat false color infrared composite; (<b>2</b>) Reference dataset manually mapped burn polygon; (<b>3</b>) European Space Agency’s (ESA’s) Fire_CCI and MCD45A1 products overlay. Despite both products show different degrees of omission and commission errors at a sub-pixel level, in our accuracy assessment, the two detections of the fire event are considered equally successful.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-12
Back to TopTop