[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Satellite-Based Assessment of Rocket Launch and Coastal Change Impacts on Cape Canaveral Barrier Island, Florida, USA
Previous Article in Journal
An Improved DETR Based on Angle Denoising and Oriented Boxes Refinement for Remote Sensing Object Detection
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
 
Journals
Remote Sensing
Volume 16
Issue 23
10.3390/rs16234422
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

First Investigation of Long-Term Methane Emissions from Wastewater Treatment Using Satellite Remote Sensing

by
Seyed Mostafa Mehrdad
1,
Bo Zhang
2,
Wenqi Guo
3,4,
Shan Du
3 and
Ke Du
1,*
1
Department of Mechanical and Manufacturing Engineering, University of Calgary, Calgary, AB T2N 1N4, Canada
2
Stantec, Atlanta, GA 30303, USA
3
Department of Computer Science, Mathematics, Physics and Statistics, University of British Columbia, Okanagan Campus, Kelowna, BC V1V 1V7, Canada
4
Weathon Software, Kelowna, BC V1X 2Z3, Canada
*
Author to whom correspondence should be addressed.
Remote Sens. 2024, 16(23), 4422; https://doi.org/10.3390/rs16234422
Submission received: 1 October 2024 / Revised: 20 November 2024 / Accepted: 22 November 2024 / Published: 26 November 2024
(This article belongs to the Section Environmental Remote Sensing)
Browse Figures
Figure 1
<p>The satellite image of the WWTP in this study.</p> "> Figure 2
<p>Flowchart diagram of data processing steps.</p> "> Figure 3
<p>Variations of average methane column concentrations inside the WWTP site from 2019 to 2023.</p> "> Figure 4
<p>Seasonally weighted average of column concentration inside the WWTP facility.</p> "> Figure 5
<p>Variation of average column concentrations at hotspots and wastewater temperature.</p> "> Figure 6
<p>Seasonal dynamics of CH4 emission locations from 2019 to 2023.</p> ">
Versions Notes

Abstract

:
Wastewater treatment (WWT) contributes 2–9% of global greenhouse gas (GHG) emissions. The noticeable uncertainty in emissions estimation is due in large part to the lack of measurement data. Several methods have recently been developed for monitoring fugitive GHG emissions from WWT. However, limited by the short duration of the monitoring, only “snapshot” data can be obtained, necessitating extrapolation of the limited data for estimating annual emissions. Extrapolation introduces substantial errors, as it fails to account for the spatial and temporal variations of fugitive emissions. This research evaluated the feasibility of studying the long-term CH4 emissions from WWT by analyzing high spatial resolution Sentinel-2 data. Satellite images of a WWT plant in Calgary, Canada, taken between 2019 and 2023, were processed to retrieve CH4 column concentration distributions. Digital image processing techniques were developed and used for extracting the time- and space-varying features of CH4 emissions, which revealed daily, monthly, seasonal, and annual variations. Emission hotspots were also identified and corroborated with ground-based measurements. Despite limitations due to atmospheric scattering, cloud cover, and sensor resolution, which affect precise ground-level concentration assessments, the findings reveal the dynamic nature of fugitive GHG emissions from WWT, indicating the need for continuous monitoring. The results also show the potential of utilizing satellite images for cost-effectively evaluating fugitive CH4 emissions.

1. Introduction

Resources invested in enhancing wastewater treatment plant (WWTP) performance have long been focused on improving the effluent quality [1,2,3,4]. In recent years, new challenges have been addressed to improve the sustainability of WWTP operation. Energy consumption and greenhouse gas (GHG) emissions are the two critical factors that an increasing number of water utilities are beginning to evaluate when planning for major capital upgrades and operational modifications [5,6]. Recent studies have identified wastewater treatment plants as important emission sources of anthropogenic GHG, contributing to climate change and air pollution [7,8]. Studies indicate that methane (CH4) emissions from municipal and industrial wastewater management account for approximately 5% of global methane production [9]. In addition, WWTPs generate and release carbon dioxide (CO2) and nitrous oxide (N2O) from the treatment processes. CO2 is also emitted during the combustion of the fossil fuels necessary for plant operation [10,11]. All of these contribute to the fugitive GHG emissions from WWTPs.
Methane is a commonly reported GHG from WWT, with a global warming potential (GWP) 25 times higher than CO2 [12]. At a typical WWTP, Daelman et al. [13] have reported that around one percent of the chemical oxygen demand (COD) that was brought into the WWTP was released as methane. CH4 emissions have been identified from various process units in WWTPs, even under toxic conditions [14,15]. Numerous studies have reported that the primary sources of methane are associated with the sludge process, including sludge holding, thickening, dewatering, anaerobic digestion, and the exhaust line from cogeneration [16,17]. The sludge process accounts for approximately 72% of methane emissions from wastewater treatment, while the remaining emissions are associated with the wastewater processes at locations where anaerobic conditions are developed and relatively higher organic loads are present [18].
Identifying and quantifying CH4 emissions is essential to the establishment of sustainable wastewater treatment practices for mitigating and reducing GHG emissions. Traditional emission factor-based estimation has been widely criticized for its lack of accuracy [19]. Recently, short-term, sporadic measurements for CH4 have been attempted in order to quantify the total emission from the entire WWTP [20]. However, at a typical WWTP, factors like weather, influent flows, pollutants loads, and operational conditions are always changing, causing the methane concentrations to fluctuate. As a result, sporadic measurements that only catch isolated “snapshots” tend to underestimate the temporal fluctuations and spatial heterogeneity of methane emissions from the WWTPs [21]. Consequently, extrapolating snapshot measurements for estimating longer-term, e.g., one year, emissions are expected to result in misleading results, which eventually will create challenges for evaluating the effectiveness of mitigation strategies.
Therefore, developing an approach for long-term monitoring provides the advantage of capturing the temporal and spatial variations of methane emissions at WWTPs. Satellite imagery has been an important tool for identifying and characterizing methane emitting facilities, such as oil and gas well pads and surface coal mines [22,23,24,25,26].
Satellite technology allows for the mapping of emission sources within a site, offering the potential to identify localized hotspots that may be disproportionately responsible for the overall emissions. Satellite data also allow the analysis of methane variations on different temporal scales, providing a more comprehensive understanding of the emission dynamics, which is critical for a detailed quantification, identification, and trending of the emissions, upon the basis of which mitigation strategies are then developed. In addition, utilizing high-frequency monitoring data means that timely detection of, and response to, unexpected emission events can be achieved [27]. Without the need for human assistance, automated detection systems powered by deep learning algorithms can identify methane emissions by analyzing multispectral satellite imagery, greatly increasing the efficiency and accuracy of the analysis [28]. Due to the various benefits of satellite imagery, this approach has been applied to the estimation of methane emissions from a range of sources, including agricultural practices such as paddy fields, landfills, and wetlands [29,30,31,32]. Although the GWP of N2O is higher than that of CH4, the use of satellite imagery to monitor nitrous oxide emissions remains a developing area of study. Nitrous oxide is a notable greenhouse gas, but its remote monitoring is complicated by its low atmospheric concentration and spectral characteristics that make it difficult to distinguish from other gases in satellite observations.
Various methods for N2O detection are under investigation, primarily using satellites equipped with hyperspectral or thermal infrared sensors capable of capturing the specific wavelengths absorbed by N2O. Satellites like Sentinel-5P and missions such as GOSAT have the spectral capability to detect and quantify trace gases, including N2O; however, they are limited by their spatial and temporal resolution [33,34]. The scarcity of data for N2O is due to its lower atmospheric concentration and weaker absorption in critical spectral bands. Additionally, N2O can interfere with other gases, complicating the analysis of satellite imagery. These limitations reduce the effectiveness of satellite imagery in investigating smaller emission sources, such as wastewater treatment plants.
The tropospheric monitoring instrument (TROPOMI) aboard the Sentinel-5 precursor satellite provides daily global methane measurements with up to 5.5 × 7 km2 pixel resolution [35]. Normally, satellite images of this resolution are considered too coarse to resolve smaller sources that are spatially clustered and produce a plume extending shorter than 1 km in distance [36]. The GHGSat microsatellite instruments are specifically designed to detect methane sources using fine pixel resolution (25–50 m) over limited domains (12 × 12 km2) and with relatively higher precision (∼1–15%) [37]. Hyperspectral imaging spectrometers designed to observe land surfaces at 1–10 nanometers spectral resolution with 30 m pixel resolution can detect large methane plumes, as was recently demonstrated with the Italian Space Agency’s PRISMA instrument [38,39]. In addition, the frequency of revisiting a specific location with targeting instruments is constrained by multiple factors, including the satellite’s spatial coverage, operational limitations regarding tasking, and the finite number of satellites available for observation. Each satellite is limited to observing a specific area during its overhead passage and, due to competing tasking priorities, imaging cannot be exclusively assigned to a single location. Furthermore, the limited number of satellites in orbit complicates the achievement of regular monitoring. A constellation of satellites is essential to increased revisit frequency, because of the improved coverage and more continuous observations across various locations.
To date, the use of satellite imagery for analyzing methane emissions from wastewater treatment has been extremely rare, if not nonexistent, for which the likely reason is that the low pixel resolution of earlier satellite images is not able to capture the complex, heterogeneous distribution of the GHG within a relatively smaller footprint. However, as previously described, recent advances in satellite technology have made significant progress meaning that high-resolution tracking capability has become available for methane monitoring. This allows for the development of image processing algorithms to capture the methane plumes from facilities of smaller footprints such as WWTPs [40].
In this study, an approach has been developed, based on Sentinel-2 images, for characterizing 5-year fugitive CH4 emissions from a wastewater treatment plant located in Canada. The analysis provides, for the first time, a comprehensive and overall mapping of fugitive methane emissions at the WWTP. The results reveal both the time-varying nature and the spatial heterogeneity of fugitive methane emissions from the wastewater treatment, indicating the importance of continuous measurements for quantifying the annual emissions for compliance reporting.

2. Literature Review

The detection and quantification of methane emissions using satellite imagery have significantly improved in recent years, driven by the need for more accurate and higher-resolution data to inform climate policy and guide mitigation efforts. Various satellite technologies and methodologies have been investigated in different studies.
Satellite-based methane detection typically relies on identifying the absorption in the short-wave infrared (SWIR) region of backscattered sunlight, specifically in the spectral regions where methane is known to absorb light. Hyperspectral satellites, with their high spectral resolution in the SWIR bands, enable a more precise determination of methane column concentrations [24,41,42]. These satellite images have either smaller spatial coverage but high spatial resolution (e.g., target-mode satellites, such as PRISMA) or low spatial resolution but broader coverage (e.g., Sentinel-5). The use of multiple Sentinel satellites in tandem for detecting and quantifying methane leaks has been covered in a recent study by Pandey et al., in which Sentinel-3 was found to enhance Sentinel-5p’s coverage to detect methane leaks of up to 8–20 tonnes per hour [30]. This study demonstrates the value of using different satellite instruments synergistically to achieve a more comprehensive characterization of methane emissions.
The study by Wecht et al. (2014) utilized a satellite-based approach for quantifying methane emissions across large regions. This research estimated that the total methane emission from U.S. livestock is approximately 40% greater than was previously estimated in the emission inventories. However, a similar discrepancy for U.S. oil and gas emissions was not observed [40]. Varon et al. (2021) have explored how Sentinel-2 data could be used to locate and quantify significant methane point sources. Using the reflectance differences of Sentinel-2 bands 11 and 12 with three retrieval methods, this research monitored methane point sources in the Hassi Messaoud Oil Field in Algeria and the Korpezhe Oil and Gas Field in Turkmenistan over several years [27].
Most recently, deep learning techniques have been integrated in to the analysis of satellite-based methane detection. Rouet-Leduc and Hulbert (2024), introduced a deep learning-based tool that overcomes the limitation of multispectral satellite data, thus enabling high-resolution methane detection. This model, once validated against airborne measurements, was found to be capable of detecting a methane point source as small as 0.01 km2, emitting approximately 200–300 kg of CH4 per hour. This approach allows for the automatic detection of methane emissions at a higher resolution, and thus improving overall monitoring capability [28]. Radman et al.’s (2023) deep learning approach involved the creation of the EfficientNet-V2L model, which was optimized to estimate the methane emission rates from Sentinel-2 data. With a Pearson R of 0.957, this approach achieved a much higher accuracy, outperforming conventional physical-based techniques such as the integrated mass enhancement (IME) method [23].
To address the limitation of hyperspecial satellites in observing small scale sources, there has been increasing interest in developing methane detection techniques using data from multispectral satellites like ESA’s Sentinel-2 constellation. Sentinel-2 provides high spatial resolution and global coverage, scanning the entire Earth every 2 to 5 days. However, multispectral satellites offer high spatial and temporal resolution at the expense of much less spectral information, with only a dozen spectral bands compared with the hundreds available on hyperspectral satellites [26,27].
Despite this, a number of studies have been recently performed using satellite imagery for methane emission quantification from industrial facilities and super emitters. Currently, there is no published research that was specifically focused on using satellite imagery to investigate methane emissions sources for smaller footprint facilities like WWTPs. Methane emission-related studies have been focused mostly on the oil and gas sector, landfills, and farming activities. Because of their higher emission rates and much larger footprint, the plumes are easier to detect using satellite imaging. In regard of WWTPs, the fugitive nature of the emissions generally creates lower intensity plumes, hence, detection and quantification can be challenging using the existing satellite data. In addition, most of the existing satellite sensors used for methane detection are incapable of achieving a spatial and temporal resolution that is high enough to capture and localize the highly heterogeneous methane emissions from wastewater treatment. Satellite remote sensing of methane provides valuable data over extensive areas by detecting specific infrared wavelengths that are absorbed by atmospheric methane. However, these measurements typically capture column-integrated concentrations from the surface to the upper atmosphere, which may not accurately represent ground-level values. Discrepancies arise from factors such as vertical methane distribution, atmospheric scattering, cloud cover, instrument calibration, and the reflectance characteristics of various surfaces. Additionally, satellites collect data at predetermined intervals and resolutions may not adequately capture small, localized sources. To mitigate these uncertainties, we focus on low-cloud days and emphasize the importance of validating satellite observations with in situ ground data to ensure precise correlation with ground-level concentrations.
This study is the first evaluation of methane emissions from WWTPs utilizing the latest satellite technologies. With the recent development and launch of advanced satellites, higher-resolution sensors have been deployed on the Sentinel-2 constellation and other specialized missions like GHGSat. The Sentinel-2 satellites were selected in this study. Sentinel-2 was originally designed for environmental risk management, land cover classification, land change detection, and terrestrial mapping. The high pixel resolution of Sentinel-2 images allows for detailed investigation of fugitive methane emissions within complex environments with smaller footprints, such as those found at WWTPs. Further- more, Sentinel-2’s frequent revisit interval provides a larger number of images, offering greater insight into the temporal variations of CH4 emissions than other satellite products.

3. Materials and Methods

3.1. Location

In this project, a WWTP in Canada was selected as the study site. It is one of the largest wastewater treatment plants in Canada.
The WWTP shown in Figure 1 operates with an average capacity of 396 million liters per day (ML/d). It employs biological nutrient removal (BNR) technology, making it not only one of the largest BNR plants in the country but also the largest cold-weather BNR plant globally. The plant must function under challenging environmental conditions, including winter air temperatures that can drop as low as −40 °C and influent temperatures that reach a minimum of 10 °C. These conditions provide a unique setting for analyzing methane emission under cold-weather operational conditions.

3.2. Satellite Image Data and Characteristics

The data for this research project were obtained from Sentinel-2 satellites. Sentinel-2 is a part of the Copernicus Programme and is an Earth observation mission that captures high-resolution optical imagery of land and coastal waters. It is operated by the European Space Agency, with the satellites built by a consortium in Germany [43,44]. The Sentinel2 constellation is composed of two satellites (S2A and S2B), positioned 180° apart in a shared sun-synchronous orbit, with a local solar time of 10:30 AM at the equator during the descending pass. Sentinel-2A was launched in June 2015, followed by Sentinel-2B in March 2017. Both satellites are equipped with a multispectral instrument (MSI) that continuously captures the Earth’s surface in 13 spectral bands, covering wavelengths from the visible to the shortwave infrared (SWIR) regions. The pixel resolution ranges from 10 to 60 m, spanning a 290 km cross-track swath [43]. An overview of the available bands of Sentinel-2 is available in Table 1.
The twin-satellite configuration of Sentinel-2 enables complete global coverage every 5 days, with revisit intervals of 2–3 days at the mid-latitudes. This study demonstrates that Sentinel-2’s SWIR bands 11 (~1560–1660 nm) and 12 (~2090–2290 nm), with a 20 m pixel resolution, are effective for detecting methane plumes from large point sources and quantifying their emission rates. These bands capture radiance within methane’s absorption features at 1650 nm and 2300 nm. Although Band 12 overlaps with the stronger and broader absorption feature at 2300 nm and is more sensitive to methane than band 11, both bands have comparable signal-to-noise ratios (SNRs) at their respective reference radiances, despite band 12 receiving lower solar irradiance [45]. Band 11 can thus serve as a proxy for the continuum, given its spectral proximity to band 12 and generally similar surface reflectance. While the multispectral instrument has a spectral resolution of approximately 100–200 nm, which is too coarse for standard hyperspectral retrieval of methane column concentrations in the SWIR, methane columns can still be estimated by analyzing reflectance differences between the spectral bands and across multiple satellite passes. To use the data sets downloaded for 5 years, preprocessing was undertaken using the semi-automatic classification (SCP) plug-in, with the default values, in QGIS software (3.28.3-Firenze), and the built-in python (version 3.9.5) console in the software [45]. The next stage of processing in this project involves applying principles from optical spectroscopy.

3.3. Methane Column Concentration Retrievals

Beer’s law is a fundamental principle in spectroscopic quantitative analysis, establishing a direct relationship between the concentration of a sample and the intensity of radiation measured by a spectrometer. This principle is essential for quantifying the concentration of a substance based on its interaction with light.
The intensity of the incident radiation was denoted as I0, referring to band 11 reflections, while the intensity of the transmitted radiation was denoted as I, which in this study was assumed to be band 12 reflections. Absorbance (A) was calculated as the logarithm of the ratio of these intensities.
A = logI0/I
To relate absorbance to the concentration of the sample, Beer’s Law—also known as the Beer–Lambert Law or Bouguer–Lambert–Beer Law, was applied. In this context,
A = ɛbC
The molar absorption coefficient (ϵ), also known as the molar absorptivity, measures how strongly a chemical species absorbs light at a specific wavelength. This coefficient is an intrinsic property of the substance. The path length (b) represents the distance the light travels through the sample and is typically measured in meters. The molar concentration (C) of the absorbing species in the sample is determined based on the absorbance and the known values of ϵ and b.
Additionally, the absorption cross-section (σ), quantifies a molecule’s ability to absorb photons. It is derived from the molar absorption coefficient and Avogadro’s number.
σ = ɛ/N
For accurate concentration determination, the absorption cross-section must be incorporated into the calculations. To achieve this, gas adsorption constants under isothermal conditions were evaluated using the Langmuir equation. The HITRAN application programming interface (HAPI) was employed to compute the integrated cross-section area of methane within band 12 wavelengths. HAPI, a python-based set of routines, provides remote access to HITRAN online data and functionality, enabling the calculation of various spectral functions, including absorption, transmittance, and radiance, based on the Hartmann–Tran (HT) profile [46]. The absorption cross-section is pivotal for these calculations, as it underlies the simulation of spectral interactions. With Equations (1)–(3), the enhancement in methane column concentration ∆Xch4 (referred to as “column concentration” hereafter) can be calculated using Equation (4).
∆ X C H 4 = − l o g I I 0 σ sec ⁡ s z a + s e c ⁡ ( v z a )
The radiative transfer model incorporates factors such as variable surface height, solar zenith angle (SZA), and instrument viewing zenith angle (VZA). For each scene, the SZA and VZA are extracted from the metadata accompanying the Sentinel-2 image tiles. This metadata provides the essential geometric information required to accurately calculate the column concentration of each pixel. Our Sentinel-2 retrievals utilize a general approach of determining methane column enhancements by examining fractional changes in the top-of-atmosphere (TOA) reflectance for band 12, when compared with a reference image where there is no, or significantly less, shortwave infrared absorption by a methane plume. By analyzing these fractional changes, methane column concentrations can be derived from any radiometric measurement proportional to TOA radiance. For this study, TOA reflectance from the Sentinel-2 Level-1C (L1C) products with cloud percentages of below 30% was used. The reason 30% percent cloudiness was selected as the threshold is to use higher quality images with the least amount of noise that derives from the presence of water vapor and clouds interacting with the light on the location. One additional criterion for processing the images was that they must all have captured the entire WWTP.
Gorroño et al. had previously validated the methane retrieval process by applying it to simulated emissions across a variety of locations, with a particular emphasis on point sources with notably higher emissions [25]. Furthermore, this retrieval method has been implemented in numerous investigations to investigate methane emissions from a variety of sources [47,48,49]. The concentration patterns from fugitive and multi-source emissions with lower emission levels at the WWTP were analyzed using this retrieval approach in our study. The use of satellite imagery for exhaustive GHG emission reporting and legislative applications could be advanced by the future validation of smaller, sporadic emission sources.
In this study, 181 images of the WWTP site over five years were gathered from 2019 to 2023. For each year, depending on the quality of the images and number of revisits, a minimum of 33 and a maximum of 44 images were processed, converting the pixel values as described to the mol per meter square column concentration. Phase one of the processing was performed on the images to determine the column concentration of the methane (mol/pixel) for each pixel using Equation (4). In the second phase of data processing, the seasonal, temporal, and annual characteristics of the dataset were investigated using image processing tools.

3.4. Seasonal and Temporal Analysis

The methane column concentration data obtained from the first phase of data processing were in the form of a tensor V with dimensions h × w × c, where h and w represent the height and width of the area of interest, respectively, and c represents different time captures. Because the intervals between captures are inconsistent, VI was generated by interpolating V along the c axis with dimensions h × w × t, where t denotes the number of days since the first capture, covering the entire period from the first to the last day of capture.
The mean value of each pixel in VI is calculated over the t axis, yielding Mi,j ∈ Rh×w. This is expressed in Equation (5).
M i , j = 1 t ∑ k = 1 t V I , i , j , k
To derive the trends of column concentration for the entire WWTP using the whole data set, TW,i ∈ Rt, is calculated by averaging VI across the h and w axes, yielding the mean emission for all pixels over time. This is expressed in Equation (6).
T W , i = 1 h × w ∑ i = 1 h ∑ k = 1 w V I , i , j , k
To calculate the seasonal average column concentration similar to the average emission of all data, the seasonal average Si was calculated by averaging the column concentration Si,j,k in each pixel (i,j) over the entire plant, Si,j,k ∈ Rh×w, where i is the season index (1–4, corresponding to spring through winter). The seasonal average map was generated by averaging each pixel’s value over the time period within the respective season. This is formally expressed in Equation (7). Di represents the days within season i.
S i , j , k = 1 D i ∑ d ∈ D i V I , i , j , k , d
A flowchart model of the entire data processing procedure is shown in Figure 2. There are two unique parts to the workflow: the first phase includes preprocessing and cleaning of the raw data, and the second phase is the analysis and interpretation of the processed data. This figure illustrates each step as well as the relationships between the various phases of data handling.

4. Results and Discussion

This study demonstrates the feasibility of obtaining long-term, high resolution methane variation within a WWTP through top-of-atmosphere observation. While the results may not precisely reflect ground-level concentrations, they reveal the heterogenous nature of source distribution. When combined with ground measurement, these findings can help identify major sources by pinpointing emissions hotspots within the WWTPs. The results illustrate the seasonal patterns in emissions and identify the locations with relatively higher methane emissions by characterizing the temporal variation, spatial distribution, and intensity of methane column concentrations from 2019 to 2023, using the data from the Sentinel-2 satellite. The findings of this study provide valuable information for identifying potential methane emission sources in order to develop efficient mitigation approaches.
Figure 3 shows the year-to-year location and strength of methane hotspots within the WWTP from 2019 to 2023. To better visualize the emission locations, the color scale was set to be 6–12, meaning that the pixels with values below 6 and above 12 are assigned the same color as for 6 and 12, respectively. The color map in Figure 4 also follows the same color scale approach. The values of column concentrations were calculated for each pixel, based on bands 11 and 12 of Sentinel-2 with a spatial resolution of 20 × 20 m (i.e., 400 m2 per pixel). Figure 3 shows that the spatial distribution of methane hotpots within the plant was similar across multiple years. Methane hotspots were primarily located in the older sludge process area, as well as the southeast and southwest areas where the activated sludge tanks and clarifiers are.
To perform a thorough evaluation of the overall dynamics of methane column concentration at the WWTP site, both seasonally weighted and non-seasonally weighted averages were calculated based on the mean column concentration of each pixel over the five-year period. Due to variations in the quantity of available data sets for each season, we implemented weighting factors to account for seasonal discrepancies. This method guaranteed that each season contributed proportionately, facilitating a more balanced and precise seasonal comparison. This analysis enabled a more detailed comparative mapping of methane emissions between seasons. For example, while the seasonally weighted and non-seasonally weighted averages were found to indicate that the locations of hotspots in the dataset were generally consistent across methods, a key difference emerged in the magnitude of the column concentrations. The weighted averages provided a more accurate representation of the strength of these hotspots.
These calculations were performed using the formulae outlined in the methodology section. It is important to note that the amount of available satellite data varied by season, being higher in spring and summer and lower in fall and winter. This discrepancy is likely due to the increased frequency of cloudy days during the cooler months, which obscure the satellite imagery and reduce data acquisition opportunities. Consequently, this seasonal variation in data availability could have biased the observed trends, particularly for methane concentration in the fall and winter seasons.
As Figure 4 indicates, the methane hotspots over the ground surface were primarily confined to the older sludge process area in the northwest and to the activated sludge tanks and clarifiers located in the southeast and southwest corners of the plant, respectively. To gain deeper insight, a thorough investigation into the processes or operations of each location could provide valuable information on how to reduce methane emissions.
To further investigate the seasonal changes in the strengths of the hotspots and emissions based on column concentrations, the seasons were categorized by month: spring (April to June), summer (July to September), fall (October to December), and winter (January to March). Please note the definition of the seasons has one month shift from that defined by air temperature, (i.e., December to February as winter). The reason for this shift is based on the consideration of the delay in temperature change in the sewage following the temperature change of the weather. In addition, the impact of temperature change on the biological process, including both the microbial communities and the metabolic activities, is also expected to take some time to be revealed. Therefore, re-defining the season allows for a realistic observation of seasonal variations in methane emission hotspots throughout the study period.
The results are presented in Figure 5, which illustrates the monthly average and 3-month moving average of column concentrations in hotspot areas in each month as well as 3-month moving average wastewater temperature. A hotspot area is defined as an area where pixels indicate that the column concentration is two standard deviations above the average value for all of the pixels over the five-years period. Identifying hotspots that vary significantly across months can indicate periods of higher-than-normal emission activities, which may correlate with operational cycles or environmental factors such as temperature and humidity. This refined method for estimating emissions provides a crucial framework for optimizing monitoring efforts.
Figure 5 illustrates the temporal variation of methane column concentrations from 2019 to 2023. Higher column concentrations tend to peak in the fall, reaching as high as 25 mol/pixel. Figure 5 reveals that methane emissions peaked approximately 2–3 months after the wastewater temperature peaked. Because methane generation is a result of microbial activities within the wastewater, the observed “lagging” of methane emissions in response to the change of the wastewater temperature is expected. A thorough interpretation of the results requires a detailed evaluation of the plant influent, effluent, and operational data, as well as the physical–chemical and biological processes, and thus is beyond the scope of this paper. The overall observation is that the temporal variations of CH4 column concentrations are positively correlated with the wastewater temperature, which is in agreement with the wastewater biological processes. During the warmer seasons when the temperatures are relatively higher, the biological activities are at a higher level, thus producing more CH4.
In addition to temporal variation, the spatial distribution of CH4 concentration over the plant site has been observed to be extremely heterogeneous and time varying. To evaluate how the CH4 emission hotspots change over time, the images were grouped by season and the CH4 strength of the images taken in the same season was averaged, as demonstrated in Figure 6. The same color scale approach as Figure 3 was used to better visualize the emission locations. Figure 6 shows that the primary hotspots in spring are located near the southeast secondary clarifiers and the older anaerobic digester and primary sludge fermenter tanks. In summer, the hotspots extend to cover more process units including the southwest activated sludge tanks and clarifiers and the northeast secondary clarifiers. The fall emission locations are similar to summer, except that additional emission sources are shown around the primary clarifiers in the northwest corner. The winter emission pattern shows a more scattered distribution of hotpots. These findings unveil the highly dynamic nature of CH4 emissions, both temporally and spatially.
These findings emphasize the necessity of conducting continuous multi-point monitoring of fugitive methane from wastewater treatment plants for capturing the temporal–spatial variation in methane emissions. In addition, the observed seasonal shifting in hotspot locations and emission intensities suggests that, when planning for mitigation measures, the emission dynamics should be considered to enhance the effectiveness.

5. Discussion and Conclusions

The results demonstrate the value of satellite remote sensing images in characterizing the spatial and temporal variation of methane emission from relatively smaller-scale industrial GHG sources like WWTPs. For the first time, this study provides a more complete, measurement-based methane emission characterization for WWTPs, revealing both the spatial and temporal variations of methane emissions.
Potential errors are introduced by factors such as atmospheric conditions, vertical methane distribution, and instrument resolution when satellite data are employed to infer surface-level methane concentrations. Satellite observations are still valuable in environmental monitoring due to their broad spatial and temporal coverage, which is not possible with ground-based measurements alone, despite the fact that these limitations influence predictive accuracy. The method can be justified in regulatory contexts by explicitly communicating these uncertainties and validating findings with ground data. Although the use of satellite data may not provide precise predictions, it effectively identifies large-scale trends and emissions anomalies, making it a powerful complementary instrument for regulatory and monitoring efforts, rather than a standalone solution.
The high spatial resolution of Sentinel-2 allows for the detection and quantification of methane column concentrations from emission sources within the plant site, enabling a detailed assessment of their environmental footprint. It is particularly useful in places where ground monitoring would either be cost prohibitive or logistically impossible to perform. Being able to monitor and reduce the emissions of small sources, like WWTPs, is of importance when seeking to meet global carbon neutrality objectives because of the large number of such types of sources [50].
By closely examining the operational dynamics at specific hotspots, potential operational inefficiencies or leaks can be identified. The findings of this study also provide regulatory implications in multiple ways. For example, instead of setting annual emission standards, the limitations can be set for different seasons, or only for the high emission months, to alleviate the costs associated with both monitoring and implementation of mitigation strategies.
Analyzing the five-year satellite imagery of methane emissions at the WWTP site provides valuable insights into the temporal and spatial dynamics of methane generation within the plant. This indicates the importance of performing continuous measurements to estimate the annual CH4 emissions.
Finally, although Sentinel-2 satellite imagery is valuable in assisting in localizing methane emission sources and estimating plumes based on the column concentrations, whether or not this approach, by itself, is capable of quantitatively evaluating the emissions still needs to be investigated. Seasonal data variability, affected by cloud cover and other atmospheric interferences, particularly impacts data availability in cooler months. These seasonal gaps may influence observed trends, potentially underrepresenting methane levels in fall and winter. Additionally, while Sentinel-2 provides high-resolution spatial data, quantitative methane estimations carry uncertainties without ground-based validation. Future research incorporating meteorological data and ground-level measurements would strengthen the reliability of satellite-based methane quantification, enabling more precise assessment and monitoring of WWTP emissions.

Author Contributions

Conceptualization, K.D., B.Z. and S.M.M.; methodology, S.M.M. and B.Z.; software, S.M.M. and W.G.; validation, K.D., S.M.M., B.Z., W.G. and S.D.; formal analysis, S.M.M. and B.Z.; investigation, S.M.M.; resources, K.D.; data curation, S.M.M.; writing—original draft preparation, S.M.M.; writing—review and editing, S.M.M., K.D. and B.Z.; visualization, S.M.M. and W.G.; supervision, K.D., S.D. and B.Z.; project administration, K.D.; funding acquisition, K.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Natural Sciences and Engineering Research Council (NSERC) of Canada, grant number: RGPIN-2020-05223.

Data Availability Statement

Data and code are available on request by contacting the authors.

Conflicts of Interest

Author Bo Zhang was employed by the company Stantec and Author Wenqi Guo was employed by the company Weathon Software. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Bolong, N.; Ismail, A.; Salim, M.R.; Matsuura, T. A review of the effects of emerging contaminants in wastewater and options for their removal. Desalination 2009, 239, 229–246. [Google Scholar] [CrossRef]
  2. Zanetti, L.; Frison, N.; Nota, E.; Tomizioli, M.; Bolzonella, D.; Fatone, F. Progress in real-time control applied to biological nitrogen removal from wastewater. A short-review. Desalination 2012, 286, 1–7. [Google Scholar] [CrossRef]
  3. Santos, A.; Ma, W.; Judd, S.J. Membrane bioreactors: Two decades of research and implementation. Desalination 2011, 273, 148–154. [Google Scholar] [CrossRef]
  4. Zhang, Q.; Hu, J.; Lee, D.-J. Aerobic granular processes: Current research trends. Bioresour. Technol. 2016, 210, 74–80. [Google Scholar] [CrossRef]
  5. Mo, W.; Zhang, Q. Energy–nutrients–water nexus: Integrated resource recovery in municipal wastewater treatment plants. J. Environ. Manag. 2013, 127, 255–267. [Google Scholar] [CrossRef]
  6. Yerushalmi, L.; Ashrafi, O.; Haghighat, F. Reductions in greenhouse gas (GHG) generation and energy consumption in wastewater treatment plants. Water Sci. Technol. 2013, 67, 1159–1164. [Google Scholar] [CrossRef]
  7. Shahabadi, M.B.; Yerushalmi, L.; Haghighat, F. Impact of process design on greenhouse gas (GHG) generation by wastewater treatment plants. Water Res. 2009, 43, 2679–2687. [Google Scholar] [CrossRef]
  8. Sweetapple, C.; Fu, G.; Butler, D. Identifying sensitive sources and key control handles for the reduction of greenhouse gas emissions from wastewater treatment. Water Res. 2014, 62, 249–259. [Google Scholar] [CrossRef]
  9. El-Fadel, M.; Massoud, M. Methane emissions from wastewater management. Environ. Pollut. 2001, 114, 177–185. [Google Scholar] [CrossRef]
  10. Zhang, B.; Mehrdad, S.; Li, S.; Du, K. To Measure or Not to Measure, that is the Question! Optical Remote Sensing of Greenhouse Gas Emissions from Wastewater Treatment. In Proceedings of the 2023 Water Environment Federation, Odors and Air Pollutants Conference, Charlotte, NC, USA, 16–19 May 2023. [Google Scholar]
  11. Li, X.; Liu, J.; Ni, P. The impact of the digital economy on CO2 emissions: A theoretical and empirical analysis. Sustainability 2021, 13, 7267. [Google Scholar] [CrossRef]
  12. Mohanakrishnan, J.; Gutierrez, O.; Meyer, R.L.; Yuan, Z. Nitrite effectively inhibits sulfide and methane production in a laboratory scale sewer reactor. Water Res. 2008, 42, 3961–3971. [Google Scholar] [CrossRef] [PubMed]
  13. Daelman, M.R.; van Voorthuizen, E.M.; van Dongen, U.G.; Volcke, E.I.; van Loosdrecht, M.C. Methane emission during municipal wastewater treatment. Water Res. 2012, 46, 3657–3670. [Google Scholar] [CrossRef] [PubMed]
  14. Wang, J.; Zhang, J.; Xie, H.; Qi, P.; Ren, Y.; Hu, Z. Methane emissions from a full-scale A/A/O wastewater treatment plant. Bioresour. Technol. 2011, 102, 5479–5485. [Google Scholar] [CrossRef] [PubMed]
  15. Aboobakar, A.; Jones, M.; Vale, P.; Cartmell, E.; Dotro, G. Methane emissions from aerated zones in a full-scale nitrifying activated sludge treatment plant. Water Air Soil Pollut. 2014, 225, 1814. [Google Scholar] [CrossRef]
  16. Yver Kwok, C.; Müller, D.; Caldow, C.; Lebègue, B.; Mønster, J.; Rella, C.; Scheutz, C.; Schmidt, M.; Ramonet, M.; Warneke, T. Methane emission estimates using chamber and tracer release experiments for a municipal waste water treatment plant. Atmos. Meas. Tech. 2015, 8, 2853–2867. [Google Scholar] [CrossRef]
  17. Oshita, K.; Okumura, T.; Takaoka, M.; Fujimori, T.; Appels, L.; Dewil, R. Methane and nitrous oxide emissions following anaerobic digestion of sludge in Japanese sewage treatment facilities. Bioresour. Technol. 2014, 171, 175–181. [Google Scholar] [CrossRef]
  18. Campos, J.L.; Valenzuela-Heredia, D.; Pedrouso, A.; Val del Río, A.; Belmonte, M.; Mosquera-Corral, A. Greenhouse Gases Emissions from Wastewater Treatment Plants: Minimization, Treatment, and Prevention. J. Chem. 2016, 2016, 3796352. [Google Scholar] [CrossRef]
  19. Paredes, M.G.; Güereca, L.P.; Molina, L.T.; Noyola, A. Methane emissions from anaerobic sludge digesters in Mexico: On-site determination vs. IPCC Tier 1 method. Sci. Total Environ. 2019, 656, 468–474. [Google Scholar] [CrossRef]
  20. Moore, D.P.; Li, N.P.; Wendt, L.P.; Castañeda, S.R.; Falinski, M.M.; Zhu, J.-J.; Song, C.; Ren, Z.J.; Zondlo, M.A. Underestimation of Sector-Wide Methane Emissions from United States Wastewater Treatment. Environ. Sci. Technol. 2023, 57, 4082–4090. [Google Scholar] [CrossRef]
  21. Du, K.; Mehrdad, S.M.; Li, S.; Zhang, B. Development of a Tiered Approach for Cost-Effectively Measuring Real-time Direct Greenhouse Gas Emissions from Wastewater Treatment. In Proceedings of the Innovations in Treatment Technology Conference, Virginia Beach, VA, USA, 21–24 May 2024. [Google Scholar]
  22. Ramachandran, N.; Irvin, J.; Omara, M.; Gautam, R.; Meisenhelder, K.; Rostami, E.; Sheng, H.; Ng, A.Y.; Jackson, R.B. Deep learning for detecting and characterizing oil and gas well pads in satellite imagery. Nat. Commun. 2024, 15, 7036. [Google Scholar] [CrossRef]
  23. Radman, A.; Mahdianpari, M.; Varon, D.J.; Mohammadimanesh, F. S2MetNet: A novel dataset and deep learning benchmark for methane point source quantification using Sentinel- 2 satellite imagery. Remote Sens. Environ. 2023, 295, 113708. [Google Scholar] [CrossRef]
  24. Irakulis-Loitxate, I.; Roger, J.; Gorroño, J.; Valverde, A.; Guanter, L. Satellites Detect Abatable Super-Emissions in One of the World’s Largest Methane Hotspot Regions. Environ. Sci. Technol. 2022, 56, 2143–2152. [Google Scholar] [CrossRef] [PubMed]
  25. Gorroño, J.; Varon, D.J.; Irakulis-Loitxate, I.; Guanter, L. Understanding the potential of Sentinel-2 for monitoring methane point emissions. Atmos. Meas. Tech. 2023, 16, 89–107. [Google Scholar] [CrossRef]
  26. Ehret, T.; De Truchis, A.; Mazzolini, M.; Morel, J.-M.; d’Aspremont, A.; Lauvaux, T.; Duren, R.; Cusworth, D.; Facciolo, G. Global Tracking and Quantification of Oil and Gas Methane Emissions from Recurrent Sentinel-2 Imagery. Environ. Sci. Technol. 2022, 56, 10517–10529. [Google Scholar] [CrossRef]
  27. Varon, D.J.; Jervis, D.; McKeever, J.; Spence, I.; Gains, D.; Jacob, D.J. High-frequency monitoring of anomalous methane point sources with multispectral Sentinel-2 satellite observations. Atmos. Meas. Tech. 2021, 14, 2771–2785. [Google Scholar] [CrossRef]
  28. Rouet-Leduc, B.; Hulbert, C. Automatic detection of methane emissions in multispectral satellite imagery using a vision transformer. Nat. Commun. 2024, 15, 3801. [Google Scholar] [CrossRef]
  29. Karimi, N.; Ng, K.T.W.; Richter, A. Prediction of fugitive landfill gas hotspots using a random forest algorithm and Sentinel-2 data. Sustain. Cities Soc. 2021, 73, 103097. [Google Scholar] [CrossRef]
  30. Pandey, S.; van Nistelrooij, M.; Maasakkers, J.D.; Sutar, P.; Houweling, S.; Varon, D.J.; Tol, P.; Gains, D.; Worden, J.; Aben, I. Daily detection and quantification of methane leaks using Sentinel-3: A tiered satellite observation approach with Sentinel-2 and Sentinel-5p. Remote Sens. Environ. 2023, 296, 113716. [Google Scholar] [CrossRef]
  31. Jang, S.; Park, J.; Lee, H.; Gou, J.; Song, I. Assessing paddy methane emissions through the identification of rice and winter crop areas using Sentinel-2 imagery in Korea. Paddy Water Environ. 2024, 22, 401–414. [Google Scholar] [CrossRef]
  32. Hondula, K.L.; DeVries, B.; Jones, C.N.; Palmer, M.A. Effects of Using High Resolution Satellite-Based Inundation Time Series to Estimate Methane Fluxes from Forested Wetlands. Geophys. Res. Lett. 2021, 48, e2021GL092556. [Google Scholar] [CrossRef]
  33. Shetty, S.; Schneider, P.; Stebel, K.; David Hamer, P.; Kylling, A.; Koren Berntsen, T. Estimating surface NO2 concentrations over Europe using Sentinel-5P TROPOMI observations and Machine Learning. Remote Sens. Environ. 2024, 312, 114321. [Google Scholar] [CrossRef]
  34. Lange, K.; Richter, A.; Burrows, J.P. Variability of nitrogen oxide emission fluxes and lifetimes estimated from Sentinel-5P TROPOMI observations. Atmos. Chem. Phys. 2022, 22, 2745–2767. [Google Scholar] [CrossRef]
  35. Hu, H.; Landgraf, J.; Detmers, R.; Borsdorff, T.; Aan de Brugh, J.; Aben, I.; Butz, A.; Hasekamp, O. Toward Global Mapping of Methane With TROPOMI: First Results Intersatellite Comparison to GOSAT. Geophys. Res. Lett. 2018, 45, 3682–3689. [Google Scholar] [CrossRef]
  36. Duren, R.M.; Thorpe, A.K.; Foster, K.T.; Rafiq, T.; Hopkins, F.M.; Yadav, V.; Bue, B.D.; Thompson, D.R.; Conley, S.; Colombi, N.K.; et al. California’s methane super-emitters. Nature 2019, 575, 180–184. [Google Scholar] [CrossRef] [PubMed]
  37. Jervis, D.; McKeever, J.; Durak, B.O.A.; Sloan, J.J.; Gains, D.; Varon, D.J.; Ramier, A.; Strupler, M.; Tarrant, E. The GHGSat-D imaging spectrometer. Atmos. Meas. Tech. 2021, 14, 2127–2140. [Google Scholar] [CrossRef]
  38. Cusworth, D.H.; Jacob, D.J.; Varon, D.J.; Chan Miller, C.; Liu, X.; Chance, K.; Thorpe, A.K.; Duren, R.M.; Miller, C.E.; Thompson, D.R.; et al. Potential of next-generation imaging spectrometers to detect and quantify methane point sources from space. Atmos. Meas. Tech. 2019, 12, 5655–5668. [Google Scholar] [CrossRef]
  39. Cusworth, D.H.; Duren, R.M.; Thorpe, A.K.; Pandey, S.; Maasakkers, J.D.; Aben, I.; Jervis, D.; Varon, D.J.; Jacob, D.J.; Randles, C.A.; et al. Multisatellite Imaging of a Gas Well Blowout Enables Quantification of Total Methane Emissions. Geophys. Res. Lett. 2021, 48, e2020GL090864. [Google Scholar] [CrossRef]
  40. Wecht, K.J.; Jacob, D.J.; Frankenberg, C.; Jiang, Z.; Blake, D.R. Mapping of North American methane emissions with high spatial resolution by inversion of SCIAMACHY satellite data. J. Geophys. Res. Atmos. 2014, 119, 7741–7756. [Google Scholar] [CrossRef]
  41. Jacob, D.J.; Varon, D.J.; Cusworth, D.H.; Dennison, P.E.; Frankenberg, C.; Gautam, R.; Guanter, L.; Kelley, J.; McKeever, J.; Ott, L.E.; et al. Quantifying methane emissions from the global scale down to point sources using satellite observations of atmospheric methane. Atmos. Chem. Phys. 2022, 22, 9617–9646. [Google Scholar] [CrossRef]
  42. Lauvaux, T.; Giron, C.; Mazzolini, M.; d’Aspremont, A.; Duren, R.; Cusworth, D.; Shindell, D.; Ciais, P. Global assessment of oil and gas methane ultra-emitters. Science 2022, 375, 557–561. [Google Scholar] [CrossRef]
  43. Drusch, M.; Del Bello, U.; Carlier, S.; Colin, O.; Fernandez, V.; Gascon, F.; Hoersch, B.; Isola, C.; Laberinti, P.; Martimort, P.; et al. Sentinel-2: ESA’s optical high-resolution mission for GMES operational services. Remote Sens. Environ. 2012, 120, 25–36. [Google Scholar] [CrossRef]
  44. Thuillier, G.; Hersé, M.; Labs, D.; Foujols, T.; Peetermans, W.; Gillotay, D.; Simon, P.; Mandel, H. The solar spectral irradiance from 200 to 2400 nm as measured by the SOLSPEC spectrometer from the ATLAS and EURECA missions. Sol. Phys. 2003, 214, 1–22. [Google Scholar] [CrossRef]
  45. Congedo, L. Semi-Automatic Classification Plugin: A Python tool for the download processing of remote sensing images in QGIS. J. Open Source Softw. 2021, 6, 3172. [Google Scholar] [CrossRef]
  46. Kochanov, R.V.; Gordon, I.E.; Rothman, L.S.; Wcisło, P.; Hill, C.; Wilzewski, J.S. HITRAN Application Programming Interface (HAPI): A comprehensive approach to working with spectroscopic data. J. Quant. Spectrosc. Radiat. Transf. 2016, 177, 15–30. [Google Scholar] [CrossRef]
  47. Tiemann, E.; Zhou, S.; Kläser, A.; Heidler, K.; Schneider, R.; Zhu, X. Machine Learning for Methane Detection and Quantification from Space—A survey. arXiv 2024, arXiv:2408.15122. [Google Scholar]
  48. Vaughan, A.; Mateo-García, G.; Gómez-Chova, L.; Růžička, V.; Guanter, L.; Irakulis-Loitxate, I. CH4Net: A deep learning model for monitoring methane super-emitters with Sentinel-2 imagery. EGUsphere 2023, 2023, 1–17. [Google Scholar] [CrossRef]
  49. Irakulis-Loitxate, I.; Roger, J.; Gorroño, J.; Valverde, A.; Guanter, L. Detection of Methane Point Sources with High-Resolution Satellites. Environ. Sci. Proc. 2023, 28, 29. [Google Scholar]
  50. Wu, Z.; Duan, H.; Li, K.; Ye, L. A comprehensive carbon footprint analysis of different wastewater treatment plant configurations. Environ. Res. 2022, 214, 113818. [Google Scholar] [CrossRef]
Remotesensing 16 04422 g001
Figure 1. The satellite image of the WWTP in this study.
Figure 1. The satellite image of the WWTP in this study.
Remotesensing 16 04422 g001
Remotesensing 16 04422 g002
Figure 2. Flowchart diagram of data processing steps.
Figure 2. Flowchart diagram of data processing steps.
Remotesensing 16 04422 g002
Remotesensing 16 04422 g003
Figure 3. Variations of average methane column concentrations inside the WWTP site from 2019 to 2023.
Figure 3. Variations of average methane column concentrations inside the WWTP site from 2019 to 2023.
Remotesensing 16 04422 g003
Remotesensing 16 04422 g004
Figure 4. Seasonally weighted average of column concentration inside the WWTP facility.
Figure 4. Seasonally weighted average of column concentration inside the WWTP facility.
Remotesensing 16 04422 g004
Remotesensing 16 04422 g005
Figure 5. Variation of average column concentrations at hotspots and wastewater temperature.
Figure 5. Variation of average column concentrations at hotspots and wastewater temperature.
Remotesensing 16 04422 g005
Remotesensing 16 04422 g006
Figure 6. Seasonal dynamics of CH4 emission locations from 2019 to 2023.
Figure 6. Seasonal dynamics of CH4 emission locations from 2019 to 2023.
Remotesensing 16 04422 g006
Table 1. Spectral band overview of Sentinel-2.
Table 1. Spectral band overview of Sentinel-2.
Band NumberWavelength and SatelliteResolution (m)
1442.7 nm (S2A), 442.3 nm (S2B)60
2492.4 nm (S2A), 492.1 nm (S2B)10
3559.8 nm (S2A), 559.0 nm (S2B)10
4664.6 nm (S2A), 665.0 nm (S2B)10
5704.1 nm (S2A), 703.8 nm (S2B)20
6740.5 nm (S2A), 739.1 nm (S2B)20
7782.8 nm (S2A), 779.7 nm (S2B)20
8832.8 nm (S2A), 833.0 nm (S2B)10
8a864.7 nm (S2A), 864.0 nm (S2B)20
9945.1 nm (S2A), 943.2 nm (S2B)60
101373.5 nm (S2A), 1376.9 nm (S2B)60
111613.7 nm (S2A), 1610.4 nm (S2B)20
122202.4 nm (S2A), 2185.7 nm (S2B)20
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mehrdad, S.M.; Zhang, B.; Guo, W.; Du, S.; Du, K. First Investigation of Long-Term Methane Emissions from Wastewater Treatment Using Satellite Remote Sensing. Remote Sens. 2024, 16, 4422. https://doi.org/10.3390/rs16234422

AMA Style

Mehrdad SM, Zhang B, Guo W, Du S, Du K. First Investigation of Long-Term Methane Emissions from Wastewater Treatment Using Satellite Remote Sensing. Remote Sensing. 2024; 16(23):4422. https://doi.org/10.3390/rs16234422

Chicago/Turabian Style

Mehrdad, Seyed Mostafa, Bo Zhang, Wenqi Guo, Shan Du, and Ke Du. 2024. "First Investigation of Long-Term Methane Emissions from Wastewater Treatment Using Satellite Remote Sensing" Remote Sensing 16, no. 23: 4422. https://doi.org/10.3390/rs16234422

APA Style

Mehrdad, S. M., Zhang, B., Guo, W., Du, S., & Du, K. (2024). First Investigation of Long-Term Methane Emissions from Wastewater Treatment Using Satellite Remote Sensing. Remote Sensing, 16(23), 4422. https://doi.org/10.3390/rs16234422

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop