[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Global Semantic Classification of Fluvial Landscapes with Attention-Based Deep Learning
Previous Article in Journal
Assessment of Post-Fire Impacts on Vegetation Regeneration and Hydrological Processes in a Mediterranean Peri-Urban Catchment
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 24
10.3390/rs16244744
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

Integrating Satellite Observations and Hydrological Models to Unravel Large TROPOMI Methane Emissions in South Sudan Wetlands

by
Yousef A. Y. Albuhaisi
1,*,
Ype van der Velde
1,
Sudhanshu Pandey
2 and
Sander Houweling
1,3
1
Department of Earth Sciences, Vrije Universiteit Amsterdam, 1081 HV Amsterdam, The Netherlands
2
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91106, USA
3
SRON Netherlands Institute for Space Research, 2333 CA Leiden, The Netherlands
*
Author to whom correspondence should be addressed.
Remote Sens. 2024, 16(24), 4744; https://doi.org/10.3390/rs16244744
Submission received: 16 October 2024 / Revised: 13 December 2024 / Accepted: 17 December 2024 / Published: 19 December 2024
(This article belongs to the Section Environmental Remote Sensing)
Browse Figures
Figure 1
<p>(<b>a</b>) South Sudan study region. Data layers from Stamen Terrain-USA/OSM and OpenStreetMap Humanitarian Data Model. (<b>b</b>) South Sudan river streams within South Sudan borders. The red dots are Hydroweb river altimetry measurement points used in the study.</p> "> Figure 2
<p>The TROPOMI XCH<sub>4</sub> annual average enhancement over the South Sudan Wetlands Region (SSWR) from 2018 to 2022 at 0.1° × 0.1° resolution. The SSWR, indicated by the black rectangle (latitude 5–10° N and longitude 28–34.5° E), encompasses the area of interest.</p> "> Figure 3
<p>Monthly CH<sub>4</sub> emissions for the South Sudan Wetlands Region (SSWR) derived from TROPOMI data from 2018 to 2022. The Y-axis represents CH<sub>4</sub> emissions anomalies in Tg CH<sub>4</sub>. The anomalies in CH<sub>4</sub> emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.</p> "> Figure 4
<p>The TROPOMI-inferred CH<sub>4</sub> enhancements (black), river altimetry (pink), GRACE equivalent water thickness (green), and ENSO index (blue and red) for the South Sudan Wetland Region (SSWR) highlighting the interconnected dynamics. The anomalies in CH<sub>4</sub> emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.</p> "> Figure 5
<p>(<b>a</b>) South Sudan Wetlands Region (SSWR) catchments, and (<b>b</b>) TROPOMI CH<sub>4</sub> emission in comparison to river altimetry measurements per catchment.</p> "> Figure 6
<p>Correlation between PCR-GLOBWB sub-surface parameters anomalies and TROPOMI CH₄ emission anomalies.</p> "> Figure 7
<p>Correlation between PCR-GLOBWB surface parameters anomalies and TROPOMI CH₄ emission anomalies.</p> "> Figure 8
<p>Correlation between PCR-GLOBWB river parameters anomalies and TROPOMI CH₄ emission anomalies.</p> "> Figure 9
<p>Comparison of normalized TROPOMI CH₄ emissions from the SSWR (black) alongside normalized water levels of Lake Victoria (green). The bars indicate the phases of the El Niño Southern Oscillation (ENSO), with El Niño events represented in blue and La Niña events in red, highlighting potential correlations between climatic phenomena and CH₄ emissions.</p> ">
Versions Notes

Abstract

:
This study presents a comprehensive investigation of Methane (CH4) emissions in the wetlands of South Sudan, employing an integrated approach that combines TROPOMI satellite data, river altimetry, and hydrological model outputs. TROPOMI data show a strong increase in CH4 concentrations over the Sudd wetlands from 2018 to 2022. We quantify CH4 emissions using these data. We find a twofold emission increase from 2018 to 2019 (9.2 ± 2.4 Tg yr−1) to 2020 to 2022 (16.3 ± 3.3 Tg yr−1). River altimetry data analysis elucidates the interconnected dynamics of river systems and CH4 emissions. We identify correlations and temporal alignments across South Sudan wetlands catchments. Our findings indicate a clear signature of ENSO driving the wetland dynamics and CH4 emissions in the Sudd by altering precipitation patterns, hydrology, and temperature, leading to variations in anaerobic conditions conducive to CH4 production. Significant correlations are found between CH4 emissions and PCR-GLOBWB-simulated soil moisture dynamics, groundwater recharge, and surface water parameters within specific catchments, underscoring the importance of these parameters on the catchment scale. Lagged correlations were found between hydrological parameters and CH4 emissions, particularly with PCR-GLOBWB-simulated capillary rise. These correlations shed light on the temporal dynamics of this poorly studied and quantified source of CH4. Our findings contribute to the current knowledge of wetland CH4 emissions and highlight the urgency of addressing the complex interplay between hydrology and carbon dynamics in these ecosystems that play a critical role in the global CH4 budget.

1. Introduction

Natural wetlands occupy approximately 8% of the Earth’s land area yet play a crucial role in the planet’s climate, serving as significant reservoirs of Soil Organic Carbon (SOC) accounting for 29% to 45% of global SOC; this organic matter accumulates under anoxic conditions in wetland soils, contributing to the estimated 1500 Pg of global soil carbon, and creating ideal conditions for microbial activity that drives methane CH4 production [1,2,3]. Given the short atmospheric lifetime of CH4 and the impact of CH4 super emitters, it has become a critical focus in global efforts to combat climate change and achieve the goals of the 2015 Paris agreement, especially as recent rapid increases in global CH4 mixing ratios underscore the urgency of addressing emissions [4]. To limit global temperature rise to below 2 °C, reducing CH4 emissions is essential, as recognized by the Global CH4 Pledge initiative of UNEP-CACC, launched at COP26; CH4 is ranked second only to carbon dioxide (CO₂) in heat-trapping potential and is responsible for approximately 16–25% of overall warming, natural sources of CH4, particularly from wetlands, remaining an area of considerable uncertainty despite significant attention to anthropogenic sources [5].
Tropical wetlands exhibit complex climate feedback mechanisms. They respond sensitively to changes in precipitation and temperature, which can alter CH4 production [6,7]. For instance, increased rainfall can lead to waterlogged conditions that promote CH4 production, while higher temperatures can accelerate microbial activity, further enhancing CH4 emissions [6]. These emissions contribute to the greenhouse gases effect, potentially intensifying climate change [8,9]. Notably, they can exhibit enhanced CH4 emissions during climate anomalies, as exemplified by the La Niña event of 2010, which led to increased CH4 emissions by 6−9 Tg yr−1 from tropical wetlands [10]. Projections, such as those by Zhang et al. [11], anticipate a significant rise in wetland CH4 emissions throughout the 21st century due to temperature-driven emission increases in tropical and high-latitude regions. While Zhang et al. [11] predict a net loss of tropical wetland areas, they also foresee an increase in boreal and Arctic wetland areas. Therefore, the sensitivity of global wetland CH4 emissions to climate fluctuations is not solely a function of temperature-driven changes but is also influenced by changes in wetland area. Thus, it is crucial to distinguish between the temperature-driven increases in emissions and the spatial redistribution of wetlands, highlighting the significant role of the tropics in contributing to global CH4 emissions under changing climatic conditions.
Despite their importance, tropical wetlands are understudied. The substantial spatial and temporal variability of wetland CH4 emissions and the logistical difficulties of on-ground monitoring make comprehensive analysis challenging. For instance, CH4 emissions can fluctuate significantly due to changes in precipitation patterns and temperature, and remote and often inaccessible locations further complicate direct observational efforts. This lack of detailed study hinders our ability to accurately quantify and predict CH4 emissions from these critical ecosystems under varying climatic conditions [12,13]. Satellite observations have demonstrated a promising potential to address these challenges [13,14]. For instance, studies by Hu et al. [15], Frankenberg et al. [16], and Pandey et al. [17] used data from TROPOMI to detect substantial CH4 enhancements over the Sudd region located in South Sudan, indicating significant emissions from the region’s wetlands. Furthermore, research by Lunt et al. [18] utilizing CH4 observations from the Japanese Greenhouse gases Observing Satellite (GOSAT), revealed that emissions from South Sudan were more than three times larger than previous estimates.
The intricate interplay between hydrology and CH4 emissions in wetlands causes the water table depth to play a critical role, as it governs the availability of oxygen—a key factor in regulating CH4 production and emissions [19]; it also influences microbial activity and the accumulation of organic matter in wetland soils [20]. Seasonal and interannual variations in wetland CH4 emissions are closely tied to shifts in water table dynamics [19]. Wetlands with fluctuating water levels, such as river–floodplain systems, release significant amounts of CH4 due to CH4 supersaturation in the water [21]. Furthermore, hydrological conditions within wetlands influence the composition and activity of microbial populations responsible for CH4 production and consumption [22].
In light of these challenges and complexities, our study builds upon the work of Pandey et al. [17] to further investigate CH4 emission dynamics within the South Sudan Wetlands Region (SSWR). Using additional years of data (2020 to 2022), we aim to study the interannual variability and identify possible trends in CH4 emissions. This approach allows us to better understand the factors influencing emissions and enhance our ability to predict future changes under varying climatic conditions. By examining climatic conditions at the catchment scale and their relationship with monthly CH4 emissions, as observed by the TROPOMI satellite, we aim to identify localized drivers of CH4 emissions within the SSWR.
In this study, we aim to investigate the relationship between river dynamics and CH4 emissions in South Sudan’s Wetlands Region (SSWR) by incorporating river stage and water level measurements grouped by catchment areas. Specifically, we seek to understand how variations in river inflows influence CH4 emission patterns, leveraging satellite data and hydrological models to provide a comprehensive analysis. Following this introduction, our study unfolds in four sections. Section 2 describes our methods and data, detailing the use of TROPOMI CH4 observations, the study area, and other relevant datasets. Section 3 analyzes CH4 emission rates by comparing data from Hydroweb river altimetry and PCR-GLOBWB model outputs for the years 2018 to 2022. In Section 4, we discuss the broader implications of our findings, while Section 5 concludes our study, highlighting key uncertainties and providing recommendations for future research in this field.

2. Materials and Methods

2.1. Study Area

The South Sudan Wetlands Region (SSWR) plays a significant role in the emission of CH4 [18] (Figure 1a). The SSWR, located along the White Nile River in South Sudan, covers an area of 9000−40,000 km2, depending on the season [18]. Hardy et al. [23] reported an increase in the SSWR area of >300% after the year 2019 using the TropWET model and concluded that the SSWR area has grown larger than 60,000 km2 and will increase further due to climate change.
Figure 1b shows the White Nile River and streams in the study area. These rivers are the SSWR’s lifeblood and impact the dynamics of the SSWR. Specifically, their water height variations are directly linked to the inundations at the SSWR [17]. This interconnected relationship between river inflow and wetland inundation patterns is an important factor controlling the CH4 emission dynamics [17].

2.2. TROPOMI CH4 Data and Emission Quantification

TROPOMI, the sole instrument onboard the Copernicus Sentinel-5 Precursor (S-5P) satellite, was launched on 13 October 2017, and operates in a sun-synchronous orbit at an altitude of 824 km [17,24]. This advanced push-broom imaging spectrometer captures spectra along a 2600 km swath while circling the Earth every 100 min, resulting in daily global coverage [17]. The data processing incorporates filtering criteria such as solar zenith angle (<70°), viewing zenith angle (<60°), and other parameters. These thresholds were chosen to minimize inaccuracies in measurements caused by extreme angles, which can introduce geometric distortions and reduce the signal-to-noise ratio in the retrieved methane concentrations [17]. Total column methane (XCH4) is retrieved from TROPOMI with consistent sensitivity within the troposphere from its absorption band at 2.3 μm, utilizing measurements of earth-reflected radiance (earthshine) recorded by the shortwave infrared (SWIR) channel of TROPOMI [15,25,26,27,28,29,30,31,32,33]. The ground pixel dimensions of TROPOMI XCH4 are 7 × 7 km2 (or 7 × 5.5 km2 since August 2019) at nadir, with larger ground pixels toward the periphery of its swath.
For this study, we employ the operational two-band retrieval product of TROPOMI (Algorithm Theoretical Baseline Document for Sentinel-5 Precursor Methane Retrieval, 2023), using the 0.76 μm O2 A and 2.3 μm CH4 bands within the near-infrared (NIR) and SWIR spectra. The retrieval of XCH4 is accomplished through the RemoTeC algorithm, using the full-physics approach to account for the effects of light path perturbations caused by atmospheric scattering due to aerosols and cirrus cloud particles [33,34,35]. We exclusively utilize high-quality XCH4 measurements obtained under favorable cloud-free conditions. XCH4 data are filtered based on the “quality assurance” (qa) flag, where (qa = 1) denotes the highest quality measurements as determined by the TROPOMI retrieval algorithm. This ensures the reliability of the data used for emission quantification (Algorithm Theoretical Baseline Document for Sentinel-5 Precursor Methane Retrieval, 2023).
To estimate emissions, we create monthly averaged TROPOMI XCH4 maps with a grid resolution of 0.1° × 0.1°, including only grid cells that have at least five high-quality TROPOMI measurements. The emissions are then calculated using the mass balance approach described by Buchwitz et al. [36] for the period spanning January 2018 to December 2022. For a more detailed explanation of the methodology used to quantify emissions, we refer to Pandey et al. [17].

2.3. Evaluation Data

2.3.1. PCR-GLOBWB Hydrological Model

The hydrological model PCRaster Global Water Balance (PCR-GLOBWB) employed in this investigation to simulate and analyze the regional hydrology operates on a grid-based framework [37,38,39]. This model was developed by the Department of Physical Geography at Utrecht University in the Netherlands [37,38,39]. PCR-GLOBWB encompasses five hydrological modules: meteorological forcing, land surface, groundwater, surface water routing, and irrigation and water use. These modules are interconnected through various exchange fluxes, such as precipitation and evaporation from soils and plant transpiration [37,38,39]. Furthermore, the model considers four primary land cover types, namely tall natural vegetation, short natural vegetation, paddy-irrigated crops, and non-paddy-irrigated crops [37,38,39]. The influence of these land cover categories on exchange fluxes is computed proportionally to their fractional grid coverage [37,38,39]. Our analysis focuses on a few selected output parameters within the land surface module (Table 1), chosen based on the literature due to their link to CH4 emissions from wetlands.
PCR-GLOBWB utilizes daily meteorological forecast data from ECMWF’s ERA5 reanalysis from 2018 to 2022, including precipitation, air temperature, and evapotranspiration information. The model is fed daily mean data with a spatial resolution of 0.1° × 0.1°. The model operates at a spatial resolution of 5 arcmin (0.08°) and processes data on a daily basis. We utilize the monthly average output from the PCR-GLOBWB model to facilitate our analysis. This choice is particularly suitable given that data from TROPOMI, Gravity Recovery and Climate Experiment (GRACE), and river altimetry are also provided on a monthly basis.

2.3.2. Water Height Measurements

Our study utilized river water level measurements from satellite altimetry from the Hydroweb database [40,41]. The Hydroweb database provides us with river water level data and serves as the primary source for monitoring water levels in the SSWR. By leveraging the power of satellite altimetry, this database provides accurate and consistent measurements, enabling us to precisely track changes in water levels [18,40,41].
In our analysis, we used water level anomalies gathered from a network of 117 measurement points distributed along key river courses (some plotted in Figure 1 to avoid stacking). These measurement points cover critical waterways such as the White Nile, Sobat, Bhar Al-ghazal, Bhar Alrab, Lol, and Chel rivers.

2.3.3. Multivariate El Niño/Southern Oscillation (ENSO) Index (MEI.v2)

ENSO MEI data were used to assess the influence of El Niño and La Niña events on high CH4 emissions from the SSWR. The ENSO MEI.v2 comprehensively represents ENSO’s intensity and phase, incorporating multiple meteorological and oceanic variables. The data are available online via (https://psl.noaa.gov/enso/mei/) (accessed on 4 June 2023).

2.3.4. Equivalent Water Height GRACE Data

To assess the relation between TROPOMI-inferred CH4 emissions and the availability of soil and groundwater, measurements from the GRACE (Gravity Recovery and Climate Experiment) satellite have been used. We used the GRACE Equivalent Water Height version 03 datasets. The data are available online via (https://grace.jpl.nasa.gov/) (accessed on 4 June 2023). The analysis involved temporal comparisons to identify correlations with CH4 emissions inferred from TROPOMI, supporting the investigation of the underlying processes.

3. Results

3.1. TROPOMI XCH4 Enhancements

We examined the changes in XCH4 mixing ratios over the SSWR. Annual average maps of TROPOMI XCH4 data from 2018 to 2022 are shown in Figure 3. Significant XCH4 enhancements are visible that align with previous remote sensing studies [15,16,17] that also reported substantial XCH4 enhancements over the SSWR. Additionally, we found distinct XCH4 enhancements over the eastern SSWR, corroborating the findings of Pandey et al. [17] and Hu et al. [15].
Table S1 provides insights into the seasonal variations in XCH4 enhancements and TROPOMI data coverage across the SSWR. TROPOMI data coverage varies seasonally, with the lowest coverage (48%) during the wet season (JJA) and the highest coverage (over 99%) during the dry season (DJF). On average over five years, data coverage is 82%, primarily due to persistent cloud cover during the wet season. The summer months of JJA show the lowest XCH4 enhancements and the highest enhancements are observed during the autumn season (SON), as indicated in Table S1.
The five-year average enhancement of XCH4 over the SSWR, as highlighted by the black rectangle in Figure 2, is 24.1 ± 4.8 ppb. This signifies a notable increase compared to the findings reported by Pandey et al. [17] for 2018 and 2019, which indicated an average enhancement of 18.8 ± 2.8 ppb.

3.2. South Sudan Emission Quantification

Our results clearly show a significant increase of CH₄ emission above the SSWR during the 5-year period from years 2018 to 2022. As shown in Table 2, the emissions increased by an order of magnitude from 2020 to 2022 to 16.3 ± 3.3 Tg yr⁻¹ compared to the period 2018–2019, almost doubling. This increase points to a change in the interactions between processes that govern emissions over time, as highlighted in the study by Pandey et al. [17].
Pandey et al. [17] showed an annual estimate by averaging the seasonal estimates, with a value of 7.4 ± 3.2 Tg yr⁻¹ for the year 2018−2019. Our result for the same period and area is higher at 9.4 ± 2.8 Tg yr⁻¹ (Table 2). The difference might be due to the version of TROPOMI XCH₄ retrievals used. We used the operational TROPOMI two-band retrieval product using a third-order polynomial function in order to model the spectral continuum variations that are necessary to minimize the effect of surface albedo variations in the retrievals.
These findings shed light on the dynamicity of CH₄ emissions in this region, while the extended dataset of TROPOMI captures an evident upward trend that thus underlines the need to consider both methodological differences and the evolving nature of the emissions when interpreting changes over time.
Figure 3 presents the monthly variation in TROPOMI-derived CH4 emissions across the SSWR expressed in units of the standard deviation of the respective time series. In all the years analyzed, a recurring pattern is evident, with emissions peaking in November, although with variations in magnitude. We observe irregular CH4 anomalies for 2020 and 2022 (See Table S1). In 2020, the annual CH4 emission reached a maximum of 17 ± 3.5 Tg yr−1 for the study period, explained by a strong emission increase from June to August. In 2022, emission was also large (15.6 ± 3.2 Tg yr−1), with a distinct peak in March and the common November peak observed in other years. In 2021, emissions followed a pattern similar to 2018 and 2019, with a stronger increase in October–November.

3.3. TROPOMI CH4 and Hydrological Data Assessment

Figure 4 displays several factors that could explain the increased TROPOMI CH4 enhancements over the SSWR. The findings in Figure 4 show TROPOMI CH4 enhancement anomalies alongside the anomalies of three other parameters: Hydroweb river altimetry, GRACE equivalent water thickness, and the ENSO variable.
The ENSO phenomenon significantly impacts the hydrological conditions in the SSWR. Between early 2018 and early 2020, El Niño brought dry conditions, which reduced river altimetry levels and water inflow, leading to lower CH4 emissions. Nevertheless, TROPOMI detected elevated CH4 levels in November 2018 and 2019, likely due to some water inflow into the SSWR; although, this inflow was relatively small compared to the larger inflows observed in the following years, from 2020 to 2022, as will be explained in the following section and Figure 5b.
The transition from El Niño to La Niña marks a significant shift in the hydrological conditions within the SSWR. Starting in mid-2020, there is a notable increase in water influx, as observed through Hydroweb river altimetry and preceding GRACE equivalent water thickness measurements (Figure 4). This surge in water influx coincides with a rapid increase in TROPOMI-inferred CH4 enhancements. The interannual variability of water inflow, represented by Hydroweb river altimetry and GRACE equivalent water thickness, closely aligns with the observed TROPOMI CH4 enhancements.
The correlation analysis of the data reveals that the correlation between TROPOMI emission and river height, as well as TROPOMI emission and the absolute ENSO index, is a moderate relationship. The following is the exact correlation between TROPOMI emission and normalized river height, which is 0.47, suggesting a moderate positive linear relationship (Figure S1). On the contrary, the correlation between normalized TROPOMI emissions with the absolute ENSO index stands at 0.31, which provides the evidence for a generally positive linear relationship, rather, weaker in nature. The result indicates the complexity of the drivers of CH4 emissions, where climate oscillations like ENSO might play a role but do not fully explain the observed variations. Overall, there is some evidence of a relationship between TROPOMI emissions and both river height and the ENSO index, but the strength of these relationships indicates that other causes of TROPOMI emission variability must be sought.

3.3.1. Catchment River Altimetry and TROPOMI CH4 Emissions

The study domain is divided into hydrological catchments to further investigate the relation between river altimetry data and TROPOMI-derived CH4 emissions (see Figure 5a). Catchments 1, 2, and 4, representing the confluence of numerous rivers traversing the SSWR, serve as the primary sources of water inflow into the SSWR. Our analysis reveals that the surges in CH4 emissions, observed by TROPOMI, align closely with the temporal variability of the river altimetry data (see Figure 5b). The emission peak at the end of 2020 coincides with a substantial change in river height in catchments 1, 2, and 4, but not in catchments 3 and 5. In 2021 and 2022, approximately at the same time of year as in 2020, the emissions also increase simultaneously with the river water level in catchments 1, 2, and 4. Some other emission peaks, notably in 2022, correlate with river height anomalies in some catchments.
The correlation coefficient between TROPOMI CH4 emissions and river altimetry measurements varies from 0.42 to 0.45 (Figure 5b). Notably, catchment 3 shows a relatively weaker correlation coefficient of 0.14. The correlation for catchment 5 is similar to those of 1, 2, and 4, but lacks the transition in variability in 2020.

3.3.2. Hydrological Model Data Assessment

The PCR-GLOBWB simulated hydrological parameters, outlined in Table 1 of Section 2.3, are used to support the interpretation of TROPOMI-derived CH4 emission variations. A Rolling-Window Analysis (RWA) method investigated lagged correlations between the PCR-GLOBWB and TROPOMI CH4 time series. From the analysis, there are generally positive lags between all PCR-GLOBWB sub-surface parameters and TROPOMI CH4 emissions, indicating that changes in hydrological parameters precede CH4 emissions, suggesting a cause–impact relationship (Figure 6 and Figure S2).
The soil moisture layers (named as upper soil storage in the PCR-GLOBWB model) in all catchments (0−5 cm and 5−30 cm layers) exhibit a range of correlations with TROPOMI CH4 emissions at less than one-month lag, further supporting the cause–impact relationship between water dynamics and CH4 emissions. In catchments 1, 3, 4, and 5, the 0−5 cm layer of upper soil storage (i.e., soil moisture) correlates well with TROPOMI CH4 emissions. Conversely, catchment 2 shows weaker correlations despite being an integral part of the SSWR. However, interannual variability aligns well with TROPOMI CH4 emissions (see Figure 6 and Figure S2). The 5−30 cm layer presents a different narrative, with all catchments showing notable correlations (see Figure 6) with a time lag of 2 to 14 days. While the 0−5 cm layer shows consistent correlations across all catchments, the soil moisture of the 5−30 cm layer exhibits more variability in correlation. Additionally, variations in this specific layer before and after 2020 partially explain increased TROPOMI CH4 emissions in the mentioned catchments (see Figure 6 and Figure S2).
Groundwater recharge parameters in catchments 1 and 5 show a strong correlation (r = 0.60 and 0.57, respectively) with TROPOMI CH4 emissions, with a time lag of 2 and 14 days, respectively. Catchment 3, centrally located within the wetland expanse, shows the highest correlation (r = 0.66) with a 14-day lag. While other catchments display considerable correlations with TROPOMI CH4 emissions, they are less pronounced than in catchments 1, 3, and 5, with time lags spanning between 14 and 30 days (see Figure 6).
The capillary rise parameter strongly correlates with TROPOMI CH4 emissions across all catchments (Figure 6). Catchment 1 shows the highest correlation (r = 0.66) with no time lag (0 days) followed by catchment 3 (r = 0.63) with an 11-day time lag. The rest of the catchments correlate well with time lags ranging from 0 to 11 days with TROPOMI CH4 emissions. However, rather than investigating the relationship between capillary rise and emissions, our analysis leaves this interesting aspect for future in-depth studies.
Surface water parameters reveal another strong correlation with TROPOMI CH4 emissions. In catchment 1, flood inundation depth shows a strong correlation (r = 0.60) with TROPOMI CH4 emissions, with a minimal time lag of 2 days (see Figure 7 and Figure S3). Other catchments show strong correlations with extended time lags of 13 to 14 days, which is less than one month. Elevated flood depth anomalies towards the end of 2020 explain the high CH4 emissions observed by TROPOMI (Figure S3). In 2021, not all catchments displayed substantial flood depth anomalies, with only catchments 4 and 5 showing corresponding anomalies (see Figure 7 and Figure S3). A notable surge in TROPOMI CH4 emissions at the beginning of 2022 is captured predominantly by catchment 2 with high flood depth anomalies.
The fraction of surface water parameter lags, ranging from 2 to 27 days, showing a good correlation for all catchments except catchment 2, where the correlation is the lowest with a time lag of 6 days (Figure 7). Near-surface soil saturation shows a different time lag for all catchments, with different correlations (Figure 7). Catchment 3 shows a strong correlation (r = 0.72), while catchment 5 has the strongest correlation with TROPOMI CH4 emissions (r = 0.51) despite being a downstream catchment (Figure 7). This distinction may relate to the hydrological behavior of the catchments forming the SSWR.
Note that Figure 7 also presents data for which TROPOMI CH4 emissions are decoupled from flood inundation depth and surface water fraction parameters. These data show a large variability in CH4 emission, even when the surface water fraction and inundation depth remain rather constant, as indicated by vertical clusters of points along the Y-axis with a small SD for the flood inundation depth and surface water fraction parameters. Here, other factors explain the emission variability, such as local conditions of the soil moisture.
The surface water level parameter shows an average time lag of 14 days for all catchments except catchment 2, with catchment 1 showing the highest correlation (r = 0.61), followed by catchment 5 (r = 0.55). Upstream catchments capture the interannual variability of TROPOMI CH4 well. Surface water anomalies at the end of 2020 align with TROPOMI CH4 emissions in subsequent seasons of 2021 and 2022 (Figure 7 and Figure S3).
The channel storage and discharge parameters (shown in Figure 8) show time lags ranging from 2 to 14 days in relation to TROPOMI CH4 emissions. Notably, catchment 1 stands out with the highest correlations, reaching r = 0.60 and r = 0.58 for channel storage and discharge, respectively (see Figure 8 and Figure S4). When examined for each catchment, these parameters show that the narrative behind the TROPOMI CH4 surges from 2020 to 2022. In the late months of 2020 to 2022, all catchments exhibited heightened anomalies, a pattern not as pronounced in the preceding years (2018 and 2019), where CH4 emissions were indeed high but comparatively less than in the subsequent years, as depicted in Figure 8. Variations across catchments directly correlate with the observed CH4 emissions during the corresponding years.

4. Discussion

The CH4 emissions from the SSWR, particularly the Sudd, have been a subject of interest in recent years. Satellite data have revealed a substantial increase in CH4 emissions from tropical Africa, including a short-term increase of 3 Tg yr−1 between 2011 and 2015 from the Sudd in the SSWR [18]. This increase is consistent with an accelerating emission increase in atmospheric CH4 during 2010–2018, with a positive trend of 1.5–2.1 Tg a−1 in the region for 2010−2016 attributed mainly to wetlands, particularly the Sudd in the SSWR [42]. Furthermore, previous estimates indicate that CH4 emissions from the SSWR, dominated by the Sudd, contribute to 13–14% (3.4 Tg yr−1) of the total emissions for East Africa [18]. Additionally, it is suggested that approximately 35–45 Tg yr−1 of the recent net growth in CH4 emissions may have been driven by natural biogenic processes, especially wetland feedback to climate change, further emphasizing the significance of wetland emissions, including those from the SSWR [18,43]
These references collectively highlight the substantial CH4 emissions from the SSWR, particularly the Sudd, and the significant contribution of these emissions to the overall CH4 budget in the region. The slight disparity in estimates (7.4 ± 3.2 Tg yr⁻¹ by Pandey et al. (2021) versus 9.2 ± 2.4 Tg yr⁻¹ in our study) may stem from the nuanced adjustments made in new TROPOMI data, specifically the utilization of a third-order polynomial function within the TROPOMI XCH4 retrieval algorithm [44].
To exclude that the interannual variation and trend in CH4 emissions that we derive from TROPOMI are influenced by uneven temporal sampling due to frequent cloud cover in the tropics, we investigated if monthly variations in data coverage correlate with the estimated CH4 emissions in this study area. Figure S5 suggests that months of higher data coverage often correspond to higher emission anomalies, though not always. CH4 emissions and data coverage show significant seasonality, as can be explained by their relationship with precipitable water. Therefore, a seasonal correlation is expected, although some influence of sampling cannot be excluded. However, the interannual variability and trend in CH4 emissions do not correlate with TROPOMI data coverage. Thus, uneven TROPOMI sampling might influence the inferred emission seasonality but cannot explain the emission increase that we found.
Regarding the data processing approach, the use of the third-order polynomial fitting method for the TROPOMI data contrasts with some prior studies that have used simpler linear models or more complex statistical methods. The third-order polynomial provides a more flexible approach, allowing us to capture non-linear relationships in the data, particularly when sudden shifts in emission patterns occur. This approach improves the fit to the data, better representing the complex dynamics of CH₄ emissions.
The use of the third-order polynomial allows for a more accurate representation of the emission trends and seasonality in the data, which would have been more difficult to achieve with simpler models. This approach ensures that the data can account for changes over time while maintaining the essential characteristics of the emission variability. Therefore, this method, particularly in the context of TROPOMI satellite data, enhances the reliability of our findings by fitting the data more closely to observed emission trends. Additionally, the flexibility of the third-order polynomial enhances our ability to capture emission increases, which was crucial in identifying the overall emission trend in this study.
The monthly emission anomalies in Figure 3 (also Table S1) reveal an intriguing temporal pattern of emissions consistently peaking in November with noteworthy anomalies in 2020 and 2022. The exceptional peak from June to August 2020 and the distinctive March peak in 2022 highlight the dynamics and complexity of CH4 emissions from the SSWR. Importantly, our findings indicate a significant increase in CH4 emissions, with an average emission of 16 ± 3.2 Tg yr−1 over the years 2020, 2021, and 2022. This increase represents a doubling compared to the emissions recorded in the years 2018 and 2019.
By examining the interannual variability of emissions, Figure 4 highlights a strong relationship between CH4 emissions and river water levels in the SSWR. This emission increase can be attributed to higher water levels, which led to an expansion of the SSWR area, as reported by Zeleke et al. [45] and Dong et al. [46]. The increased extent of the SSWR area directly influenced CH4 emissions through enhanced wetland conditions conducive to methanogenesis. Furthermore, Pandey et al. [17] found that the seasonality of CH4 emissions is primarily driven by the inundation extent of the modeled CH4 emissions by the LPJ-wsl and Wetcharts process models. Furthermore, the GRACE equivalent water thickness data (Figure 4) also exhibit similar trends consistent with river water levels, though an early increase in 2020 might be linked to human activities or water management practices. Additionally, rising water levels in Lake Victoria (Figure 9)—the primary inflow source for the White Nile—is likely a significant factor influencing the observed CH4 emissions trend in the SSWR. These increased levels in Lake Victoria have been attributed to both natural factors, such as increased precipitation, and anthropogenic influences, including dam-regulated outflows [47]. Such changes, modulated by ENSO events, can create cascading effects on the SSWR, where larger water inflows expand wetland areas, fostering favorable conditions for CH4 production through enhanced methanogenesis. The rise in Lake Victoria water levels aligns with trends in the SSWR, confirming a hydrological connection between both.
In this study, we did not directly analyze the spatial extent of the SSWR wetlands; rather, we inferred that substantial inflows into the SSWR region likely contributed to its areal expansion. Our findings align with previous studies, reinforcing the link between seasonal inundation and CH₄ emissions in this region. For example, Hardy et al. [23] investigated the SSWR role in CH₄ emissions using satellite data to monitor seasonal changes in water coverage and vegetation, identifying CH₄ emission hotspots during peak flooding periods. Their study highlighted the October–November–December (OND) season as having the highest flood levels, especially in November, which aligns with the elevated emissions we observed in November, as shown in Figure 4. Furthermore, Hardy et al. [23] calculated a significant increase in the SSWR areal extent from 2019 to 2023, nearly tripling in size compared to pre-2019 levels. This expansion further supports our observation of heightened TROPOMI CH₄ emissions over the SSWR.
Recent studies have underscored the crucial role ENSO plays in influencing wetland CH4 emissions, particularly from the SSWR [48]. The impact of ENSO on CH4 emissions can be due to variations in temperature as well as precipitation [10]. While temperature fluctuations can influence CH4 release, especially during ENSO events, tropical wetlands are more sensitive to changes in water levels and surface water dynamics due to the relatively stable temperatures in these regions [49]. Our results in Figure 4 and Figure 9 clearly aligned with the ENSO phases. During the El Niño period, data on river water levels, GRACE, and Lake Victoria indicate low water inflow to the SSWR, resulting in lower emissions observed in 2018 and 2019. Our results show that the emissions increased in 2020, 2021, and 2022 as the region shifted to the La Niña phase. However, tracking monthly changes within each year remains challenging because the SSWR is treated as a single rectangular unit, which may overlook important details about how the wetlands respond to climatological variations.
To enhance our understanding of SSWR emissions further, a catchment-based hydrological approach has been applied. By analyzing river height data and TROPOMI CH4 emissions for each catchment within the SSWR (Figure 5), the notable increase in emissions toward the end of 2020 aligns with rising water levels in catchments 1, 2, and 4 as they emerge as key water sources for the wetlands, extending beyond the geographic boundaries of the SSWR. Notably, all catchments used in this study are subject to unique climatological variation, including fluctuations in precipitation, temperature, evapotranspiration, and ENSO-related effect, which result in distinct hydrological responses, as evident in the differences in river water levels across catchments. For example, as shown in Figure 5b, the high emissions observed by TROPOMI during 2018 and 2019 correspond well with river height data from catchments 1, 3, and 5, while the inflow anomalies from catchments 2 and 4 were weak during this period. Conversely, the high emissions observed from 2020 onward align better with the strong inflow anomalies from catchments 2 and 4, which were recently captured by catchments 1 and 5 while absent in catchment 3. This helps explain the high CH₄ emissions observed in November 2018 and 2019, even in the absence of significant inflow from the upstream catchments. This aligns with the earlier findings that local wetland dynamics, such as water coverage and vegetation, play a critical role in CH4 production [18,23]. This temporal and spatial variation in hydrological influence underscores the importance of considering multiple inflow sources when analyzing CH4 emissions from the SSWR.
Recent studies have documented considerable changes in the extent of wetlands in the SSWR, primarily due to climate variability, hydrological dynamics, and human activities. For instance, Lunt et al. [18] used satellite data to show that wetland extent in the SSWR, particularly in the Sudd, has increased significantly in response to intensified inflows from the White Nile. This increase is often linked to seasonal changes in precipitation and river inflow, as well as broader climatic factors such as the ENSO that influence the region’s hydrology, which supports our findings regarding the influence of ENSO on the SSWR. However, the full picture is even more complex with an important role of regional catchments influencing the SSWR hydrology.
Dong et al. [46] developed a statistical model using a comprehensive observational inundation dataset from 2018 to 2022, combined with satellite-derived water storage and precipitation data across the Sudd wetland. This updated wetland extent data was then applied to enhance CH₄ emission estimates from the wetland. The study reported a considerable increase in the total wetland areas from 2019 to 2022 driven by the anomalous precipitation in the upstream area. Similarly, Gumbricht et al. [50] mapped wetland areas and highlighted the substantial seasonal fluctuations in tropical wetland extent due to several hydrological factors. Pandey et al. [17] found that models estimating inundation based solely on local precipitation tend to underestimate wetland extents, as they do not account for the influence of river inflow. These studies collectively underscore the dynamic nature of wetland boundaries in the SSWR, emphasizing the need to consider long-range river transport as well as regional catchment-inflow-driven hydrological processes, both influenced by climate variations, when assessing CH₄ emissions and wetland impacts on regional ecology.
Pandey et al. [17] discussed a lag between the peak CH4 emissions observed by TROPOMI and the timing of wetland inundations. This lag occurs because CH4 emissions typically follow a delayed response to changes in water levels and vegetation growth [51]. Inundations create anaerobic (low-oxygen) conditions in wetland soils, promoting CH4 production, but this process takes time to reach its peak [51]. Pandey et al. [17] reported that the CH4 emissions measured by TROPOMI peak approximately one month after the actual inundation event, reflecting the time needed for microbial activity to generate CH4 once the wetland is flooded. Pandey et al. [17] and other researchers added that this lag is important for understanding the timing and magnitude of CH4 emissions from wetlands and could influence how satellite data are interpreted in relation to seasonal hydrological changes. Therefore, we delved into the hydrology of wetlands by testing a list of hydrological parameters (Section 2.3.1) that have an influence on the production of CH4.
The integration of PCR-GLOBWB model parameters (described in Section 2.3) with TROPOMI CH4 emissions is used to increase the robustness and credibility of our analyses. The use of PCR-GLOBWB model parameters revealed a range of time lags between hydrological parameters and TROPOMI CH4 emissions. Previous studies, such as Jerman et al. [51], have indicated that CH4 production typically begins after a lag of 84 days at 15 °C and a minimum of 7 days at 37 °C—the optimum temperature for methanogenesis. Our results align with these findings, falling within the specified range. However, for some parameters in specific catchments, the lag was generally shorter than 7 days. Interestingly, some hydrological parameters, such as capillary rise in catchment 1, showed zero-days lag times with TROPOMI CH4 emissions (Figure S2). Additionally, most tested hydrological parameters displayed varied lag times across different catchments, suggesting that each catchment’s CH4 emissions are driven independently, responding uniquely when optimal conditions for CH4 production arise, as observed in our analysis. Therefore, to understand the temporal variability of CH4 emissions from the SSWR, it is important to consider the temporal dynamics of its hydrological catchments. Future research is needed to explore these influences in further detail.

5. Conclusions

Our comprehensive analysis of TROPOMI XCH4 enhancements, South Sudan Wetlands Region (SSWR) emission quantification, catchment river altimetry, and PCR-GLOBWB hydrological model data provides valuable insights into the complex dynamics of CH4 emissions in the SSWR. Our findings align with previous remote sensing studies, indicating substantial XCH4 enhancements over the Sudd wetlands. The quantification of emissions reveals a remarkable twofold increase in CH4 emissions from 2018 to 2019 to 2020 to 2022. The integration of river altimetry data showcases the interconnectedness of river systems and CH4 emissions, with notable correlations and temporal alignments across catchments. The impact of ENSO on wetland dynamics and CH4 emissions is evident, highlighting the influence of climatic variability on this critical ecosystem. A remarkably positive significant correlation of the capillary rise parameter across catchments points to one promising direction of further research into this subject, with a view to consider further complex interactions between hydrological parameters and CH₄ emissions. Our findings contribute to the broader understanding of wetland dynamics and the factors influencing CH4 emissions. The temporal patterns revealed in our study, the impact of climatic events, and the localized nature of hydrological influences underscore the need for catchment-specific analyses in wetland research. As wetlands continue to play a crucial role in the global CH4 budget, our study highlights the importance of addressing the complex interplay between hydrology and carbon dynamics.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/rs16244744/s1.

Author Contributions

Conceptualization, Y.A.Y.A. and S.H.; methodology, Y.A.Y.A.; software, Y.A.Y.A. and S.P.; validation, Y.A.Y.A. and S.H.; formal analysis, Y.A.Y.A.; investigation, Y.A.Y.A.; resources, Y.A.Y.A.; data curation, Y.A.Y.A.; writing—original draft preparation, Y.A.Y.A.; writing—review and editing, S.H., Y.v.d.V. and S.P.; visualization, Y.A.Y.A.; supervision, S.H.; project administration, Y.A.Y.A.; funding acquisition, S.H. All authors have read and agreed to the published version of the manuscript.

Funding

The project is funded by the VU Amsterdam, under the carbon cycle data assimilation in modeling CH4 emissions from natural wetlands (project no. 2922502).

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

We express our gratitude to the PCR-GLOBWB team for their valuable guidance and assistance in utilizing the model. Special thanks to our reviewers for providing constructive comments and insightful suggestions. We acknowledge the Principal Investigators of the Hydroweb, GRACE, and NOAA datasets used in this study for generously sharing these resources with the research community. Additionally, our appreciation goes to SURFSara for granting access to the Snellius HPC platform through computing grant no. 17235, enabling the computational aspects of our research. Part of this work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under contract to the National Aeronautics and Space Administration (80NM0018D0004).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Chen, H.; Xu, X.; Fang, C.; Li, B.; Nie, M. Differences in the Temperature Dependence of Wetland CO2 and CH4 Emissions Vary with Water Table Depth. Nat. Clim. Change 2021, 11, 766–771. [Google Scholar] [CrossRef]
  2. Mitsch, W.J.; Bernal, B.; Nahlik, A.M.; Mander, Ü.; Zhang, L.; Anderson, C.J.; Jørgensen, S.E.; Brix, H. Wetlands, Carbon, and Climate Change. Landsc. Ecol. 2013, 28, 583–597. [Google Scholar] [CrossRef]
  3. Nahlik, A.M.; Fennessy, M.S. Carbon Storage in US Wetlands. Nat. Commun. 2016, 7, 13835. [Google Scholar] [CrossRef] [PubMed]
  4. IPCC Summary for Policymakers: Synthesis Report. In Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; IPCC: Geneva, Switzerland, 2023; pp. 1–34. [CrossRef]
  5. Rosentreter, J.A.; Borges, A.V.; Deemer, B.R.; Holgerson, M.A.; Liu, S.; Song, C.; Melack, J.; Raymond, P.A.; Duarte, C.M.; Allen, G.H.; et al. Half of Global Methane Emissions Come from Highly Variable Aquatic Ecosystem Sources. Nat. Geosci. 2021, 14, 225–230. [Google Scholar] [CrossRef]
  6. Shaw, J.T.; Allen, G.; Barker, P.; Pitt, J.R.; Pasternak, D.; Bauguitte, S.J.-B.; Lee, J.; Bower, K.N.; Daly, M.C.; Lunt, M.F.; et al. Large Methane Emission Fluxes Observed From Tropical Wetlands in Zambia. Glob. Biogeochem. Cycles 2022, 36, e2021GB007261. [Google Scholar] [CrossRef]
  7. Gerlein-safdi, C.; Bloom, A.A.; Plant, G.; Kort, E.A.; Ruf, C.S. Improving Representation of Tropical Wetland Methane Emissions With CYGNSS Inundation Maps. Glob. Biogeochem. Cycles 2021, 35, e2020GB006890. [Google Scholar] [CrossRef]
  8. Villa, J.A. Functional Representation of Biological Components in Methane-Cycling Processes in Wetlands Improves Modeling Predictions. J. Geophys. Res. Biogeosci. 2020, 125, e2020JG005794. [Google Scholar] [CrossRef]
  9. Sheng, J.X.; Jacob, D.J.; Turner, A.J.; Maasakkers, J.D.; Sulprizio, M.P.; Bloom, A.A.; Andrews, A.E.; Wunch, D. High-Resolution Inversion of Methane Emissions in the Southeast US Using SEAC4RS Aircraft Observations of Atmospheric Methane: Anthropogenic and Wetland Sources. Atmos. Chem. Phys. 2018, 18, 6483–6491. [Google Scholar] [CrossRef]
  10. Pandey, S.; Houweling, S.; Krol, M.; Aben, I.; Monteil, G.; Nechita-Banda, N.; Dlugokencky, E.J.; Detmers, R.; Hasekamp, O.; Xu, X.; et al. Enhanced Methane Emissions from Tropical Wetlands during the 2011 La Niña. Sci. Rep. 2017, 7, 45759. [Google Scholar] [CrossRef] [PubMed]
  11. Zhang, Z.; Zimmermann, N.E.; Stenke, A.; Li, X.; Hodson, E.L.; Zhu, G.; Huang, C.; Poulter, B. Emerging Role of Wetland Methane Emissions in Driving 21st Century Climate Change. Proc. Natl. Acad. Sci. USA 2017, 114, 9647–9652. [Google Scholar] [CrossRef] [PubMed]
  12. Albuhaisi, Y.A.Y.; van der Velde, Y.; Houweling, S. The Importance of Spatial Resolution in the Modeling of Methane Emissions from Natural Wetlands. Remote Sens. 2023, 15, 2840. [Google Scholar] [CrossRef]
  13. Petrescu, A.M.R.; Lohila, A.; Tuovinen, J.-P.; Baldocchi, D.D.; Desai, A.R.; Roulet, N.T.; Vesala, T.; Dolman, A.J.; Oechel, W.C.; Marcolla, B.; et al. The Uncertain Climate Footprint of Wetlands Under Human Pressure. Proc. Natl. Acad. Sci. USA 2015, 112, 4594–4599. [Google Scholar] [CrossRef] [PubMed]
  14. Albuhaisi, Y.A.Y.; van der Velde, Y.; De Jeu, R.; Zhang, Z.; Houweling, S. High-Resolution Estimation of Methane Emissions from Boreal and Pan-Arctic Wetlands Using Advanced Satellite Data. Remote Sens. 2023, 15, 3433. [Google Scholar] [CrossRef]
  15. 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 and Intersatellite Comparison to GOSAT. Geophys. Res. Lett. 2018, 45, 3682–3689. [Google Scholar] [CrossRef]
  16. Frankenberg, C.; Aben, I.; Bergamaschi, P.; Dlugokencky, E.J.; Van Hees, R.; Houweling, S.; Van Der Meer, P.; Snel, R.; Tol, P. Global Column-Averaged Methane Mixing Ratios from 2003 to 2009 as Derived from SCIAMACHY: Trends and Variability. J. Geophys. Res. Atmos. 2011, 116, 1–12. [Google Scholar] [CrossRef]
  17. Pandey, S.; Houweling, S.; Lorente, A.; Borsdorff, T.; Tsivlidou, M.; Anthony Bloom, A.; Poulter, B.; Zhang, Z.; Aben, I. Using Satellite Data to Identify the Methane Emission Controls of South Sudan’s Wetlands. Biogeosciences 2021, 18, 557–572. [Google Scholar] [CrossRef]
  18. Lunt, M.M.; Palmer, P.P.; Feng, L.; Taylor, C.C.; Boesch, H.; Parker, R.R. An Increase in Methane Emissions from Tropical Africa between 2010 and 2016 Inferred from Satellite Data. Atmos Chem Phys 2019, 19, 14721–14740. [Google Scholar] [CrossRef]
  19. Fisher, R.; Lowry, D.; Lanoisellé, M.; Brownlow, R.; Pyle, J.A.; Cain, M.; Warwick, N.; Skiba, U.; Skiba, U.; Drewer, J.; et al. Measurement of the 13C Isotopic Signature of Methane Emissions From Northern European Wetlands. Glob. Biogeochem. Cycles 2017, 31, 605–623. [Google Scholar] [CrossRef]
  20. Methane Cycling in Wetland Ecosystems; Vrije Universiteit Amsterdam: Amsterdam, The Netherlands, 2024.
  21. Cui, M.; Ma, A.; Qi, H.; Bai, Z.; Zhuang, G.; Zhao, G. Warmer Temperature Accelerates Methane Emissions From the Zoige Wetland on the Tibetan Plateau Without Changing Methanogenic Community Composition. Sci. Rep. 2015, 5, 11616. [Google Scholar] [CrossRef] [PubMed]
  22. Koffi, E.N.; Bergamaschi, P.; Alkama, R.; Cescatti, A. An Observation-Constrained Assessment of the Climate Sensitivity and Future Trajectories of Wetland Methane Emissions. Sci. Adv. 2020, 6, eaay4444. [Google Scholar] [CrossRef]
  23. Hardy, A.; Palmer, P.I.; Oakes, G. Satellite Data Reveal How Sudd Wetland Dynamics Are Linked with Globally-Significant Methane Emissions. Environ. Res. Lett. 2023, 18, 074044. [Google Scholar] [CrossRef]
  24. Veefkind, J.P.; Aben, I.; McMullan, K.; Förster, H.; de Vries, J.; Otter, G.; Claas, J.; Eskes, H.J.; de Haan, J.F.; Kleipool, Q.; et al. TROPOMI on the ESA Sentinel-5 Precursor: A GMES Mission for Global Observations of the Atmospheric Composition for Climate, Air Quality and Ozone Layer Applications. Remote Sens. Environ. 2012, 120, 70–83. [Google Scholar] [CrossRef]
  25. Varon, D.J.; Jacob, D.J.; McKeever, J.; Jervis, D.; Durak, B.O.A.; Xia, Y.; Huang, Y. Quantifying Methane Point Sources From Fine-Scale Satellite Observations of Atmospheric Methane Plumes. Atmos. Meas. Tech. 2018, 11, 5673–5686. [Google Scholar] [CrossRef]
  26. van Geffen, J.; Boersma, K.F.; Eskes, H.; Sneep, M.; ter Linden, M.; Zara, M.; Veefkind, J.P. S5P/TROPOMI NO2 Slant Column Retrieval: Method, Stability, Uncertainties, and Comparisons Against OMI. 2019. Available online: https://amt.copernicus.org/preprints/amt-2019-471/amt-2019-471.pdf (accessed on 4 June 2023). [CrossRef]
  27. Dobber, M.; Kleipool, Q.; Veefkind, P.; Levelt, P.F.; Rozemeijer, N.M.; Hoogeveen, R.W.; Aben, I.; de Vries, J.; Otter, G. From Ozone Monitoring Instrument (OMI) to Tropospheric Monitoring Instrument (TROPOMI). In Proceedings of the International Conference on Space Optics 2008, Toulouse, France, 14–17 October 2008. [Google Scholar] [CrossRef]
  28. Liu, M.; Van der A, R.; van Weele, M.; Eskes, H.; Lu, X.; Veefkind, P.; de Laat, J.; Kong, H.; Wang, J.; Sun, J.; et al. A New Divergence Method to Quantify Methane Emissions Using Observations of Sentinel-5p TROPOMI. ESS Open Archive 2021, 48. [Google Scholar] [CrossRef]
  29. Reuter, M.; Buchwitz, M.; Schneising, O.; Krautwurst, S.; O’Dell, C.; Richter, A.; Bovensmann, H.; Burrows, J.P. Towards monitoring localized CO2 emissions from space: Co-located regional CO2 and NO2 enhancements observed by the OCO-2 and S5P satellites. Atmos. Chem. Phys. 2019, 19, 9371–9383. [Google Scholar] [CrossRef]
  30. Schneising, O.; Buchwitz, M.; Reuter, M.; Bovensmann, H.; Burrows, J.P.; Borsdorff, T.; Deutscher, N.M.; Feist, D.G.; Griffith, D.; Hase, F.; et al. A Scientific Algorithm to Simultaneously Retrieve Carbon Monoxide and Methane From TROPOMI Onboard Sentinel-5 Precursor. Atmos. Meas. Tech. 2019, 12, 6771–6802. [Google Scholar] [CrossRef]
  31. Maasakkers, J.D.; Varon, D.J.; Elfarsdóttir, A.; McKeever, J.; Jervis, D.; Mahapatra, G.; Pandey, S.; Lorente, A.; Borsdorff, T.; Foorthuis, L.R.; et al. Using Satellites to Uncover Large Methane Emissions From Landfills. Sci. Adv. 2022, 8, eabn9683. [Google Scholar] [CrossRef]
  32. Qu, Z.; Jacob, D.J.; Shen, L.; Lu, X.; Zhang, Y.; Scarpelli, T.R.; Nesser, H.; Sulprizio, M.P.; Maasakkers, J.D.; Bloom, A.A.; et al. Global Distribution of Methane Emissions: A Comparative Inverse Analysis of Observations From the TROPOMI and GOSAT Satellite Instruments. Atmos. Chem. Phys. 2021, 21, 14159–14175. [Google Scholar] [CrossRef]
  33. Hu, H.; Hasekamp, O.; Butz, A.; Galli, A.; Landgraf, J.; Aan De Brugh, J.; Borsdorff, T.; Scheepmaker, R.; Aben, I. The Operational Methane Retrieval Algorithm for TROPOMI. Atmos. Meas. Tech. 2016, 9, 5423–5440. [Google Scholar] [CrossRef]
  34. Hasekamp, O.; Lorente, A.; Hu, H.; Butz, A.; Aan De Brugh, J.; Landgraf, J. Algorithm Theoretical Baseline Document for Sentinel-5 Precursor Methane Retrieval. Available online: https://sentinel.esa.int/web/sentinel/home (accessed on 4 June 2023).
  35. Butz, A.; Galli, A.; Hasekamp, O.; Landgraf, J.; Tol, P.; Aben, I. TROPOMI Aboard Sentinel-5 Precursor: Prospective Performance of CH 4 Retrievals for Aerosol and Cirrus Loaded Atmospheres. Remote Sens. Environ. 2012, 120, 267–276. [Google Scholar] [CrossRef]
  36. Buchwitz, M.; Schneising, O.; Reuter, M.; Heymann, J.; Krautwurst, S.; Bovensmann, H.; Burrows, J.P.; Boesch, H.; Parker, R.J.; Somkuti, P.; et al. Satellite-Derived Methane Hotspot Emission Estimates Using a Fast Data-Driven Method. Atmos. Chem. Phys. 2017, 17, 5751–5774. [Google Scholar] [CrossRef]
  37. Bierkens, M.F.P. Global Hydrology 2015: State, Trends, and Directions. Water Resour. Res. 2015, 51, 4923–4947. [Google Scholar] [CrossRef]
  38. de Graaf, I.E.M.; Sutanudjaja, E.; van Beek, L.P.H.; Bierkens, M.F.P. A High-Resolution Global-Scale Groundwater Model. Hydrol. Earth Syst. Sci. 2015, 19, 823–837. [Google Scholar] [CrossRef]
  39. Sutanudjaja, E.; van Beek, L.P.H.; Wanders, N.; Wada, Y.; Bosmans, J.; Drost, N.; van der Ent, R.; de Graaf, I.; Hoch, J.M.; de Jong, K.; et al. PCR-GLOBWB 2: A 5 Arcmin Global Hydrological and Water Resources Model. Geosci. Model. Dev. 2018, 11, 2429–2453. [Google Scholar] [CrossRef]
  40. Santos da Silva, J.; Calmant, S.; Seyler, F.; Rotunno Filho, O.C.; Cochonneau, G.; Mansur, W.J. Water Levels in the Amazon Basin Derived from the ERS 2 and ENVISAT Radar Altimetry Missions. Remote Sens. Environ. 2010, 114, 2160–2181. [Google Scholar] [CrossRef]
  41. Crétaux, J.F.; Arsen, A.; Calmant, S.; Kouraev, A.; Vuglinski, V.; Bergé-Nguyen, M.; Gennero, M.C.; Nino, F.; Abarca Del Rio, R.; Cazenave, A.; et al. SOLS: A Lake Database to Monitor in the Near Real Time Water Level and Storage Variations from Remote Sensing Data. Adv. Space Res. 2011, 47, 1497–1507. [Google Scholar] [CrossRef]
  42. Zhang, Y.; Jacob, D.J.; Lu, X.; Maasakkers, J.D.; Scarpelli, T.R.; Sheng, J.X.; Shen, L.; Qu, Z.; Sulprizio, M.P.; Chang, J.; et al. Attribution of the Accelerating Increase in Atmospheric Methane during 2010-2018 by Inverse Analysis of GOSAT Observations. Atmos. Chem. Phys. 2021, 21, 3643–3666. [Google Scholar] [CrossRef]
  43. Nisbet, E.G.; Manning, M.R.; Dlugokencky, E.J.; Michel, S.E.; Lan, X.; Röckmann, T.; Denier van der Gon, H.A.C.; Schmitt, J.; Palmer, P.I.; Dyonisius, M.N.; et al. Atmospheric Methane: Comparison Between Methane’s Record in 2006–2022 and During Glacial Terminations. Glob. Biogeochem. Cycles 2023, 37, e2023GB007875. [Google Scholar] [CrossRef]
  44. Lorente, A.; Borsdorff, T.; Martinez-Velarte, M.C.; Landgraf, J. Accounting for Surface Reflectance Spectral Features in TROPOMI Methane Retrievals. Atmos. Meas. Tech. 2023, 16, 1597–1608. [Google Scholar] [CrossRef]
  45. Zeleke, T.T.; Zakaria Wani Lukwasa, A.; Ture Beketie, K.; Yayeh Ayal, D. Analysis of Spatio-Temporal Precipitation and Temperature Variability and Trend over Sudd-Wetland, Republic of South Sudan. Clim. Serv. 2024, 34, 100451. [Google Scholar] [CrossRef]
  46. Dong, B.; Peng, S.; Liu, G.; Pu, T.; Gerlein-Safdi, C.; Prigent, C.; Lin, X. Underestimation of Methane Emissions From the Sudd Wetland: Unraveling the Impact of Wetland Extent Dynamics. Geophys. Res. Lett. 2024, 51, e2024GL110690. [Google Scholar] [CrossRef]
  47. Siderius, C.; Van Walsum, P.E.V.; Roest, C.W.J.; Smit, A.A.M.F.R.; Hellegers, P.J.G.J.; Kabat, P.; Van Ierland, E.C. The Role of Rainfed Agriculture in Securing Food Production in the Nile Basin. Environ. Sci. Policy 2016, 61, 14–23. [Google Scholar] [CrossRef]
  48. Rhodes, R.H.; Brook, E.J.; McConnell, J.R.; Blunier, T.; Sime, L.C.; Faïn, X.; Mulvaney, R. Atmospheric Methane Variability: Centennial-Scale Signals in the Last Glacial Period. Glob. Biogeochem. Cycles 2017, 31, 575–590. [Google Scholar] [CrossRef]
  49. Bloom, A.A.; Palmer, P.I.; Fraser, A.; Reay, D.S.; Frankenberg, C. Large-Scale Controls of Methanogenesis Inferred from Methane and Gravity Spaceborne Data. Science 2010, 327, 322–325. [Google Scholar] [CrossRef]
  50. Gumbricht, T.; Roman-Cuesta, R.M.; Verchot, L.; Herold, M.; Wittmann, F.; Householder, E.; Herold, N.; Murdiyarso, D. An Expert System Model for Mapping Tropical Wetlands and Peatlands Reveals South America as the Largest Contributor. Glob. Chang. Biol. 2017, 23, 3581–3599. [Google Scholar] [CrossRef]
  51. Jerman, V.; Metje, M.; Mandić-Mulec, I.; Frenzel, P. Wetland restoration and methanogenesis: The activity of microbial populations and competition for substrates at different temperatures. Biogeosciences 2009, 6, 1127–1138. [Google Scholar] [CrossRef]
Remotesensing 16 04744 g001
Figure 1. (a) South Sudan study region. Data layers from Stamen Terrain-USA/OSM and OpenStreetMap Humanitarian Data Model. (b) South Sudan river streams within South Sudan borders. The red dots are Hydroweb river altimetry measurement points used in the study.
Figure 1. (a) South Sudan study region. Data layers from Stamen Terrain-USA/OSM and OpenStreetMap Humanitarian Data Model. (b) South Sudan river streams within South Sudan borders. The red dots are Hydroweb river altimetry measurement points used in the study.
Remotesensing 16 04744 g001
Remotesensing 16 04744 g002
Figure 2. The TROPOMI XCH4 annual average enhancement over the South Sudan Wetlands Region (SSWR) from 2018 to 2022 at 0.1° × 0.1° resolution. The SSWR, indicated by the black rectangle (latitude 5–10° N and longitude 28–34.5° E), encompasses the area of interest.
Figure 2. The TROPOMI XCH4 annual average enhancement over the South Sudan Wetlands Region (SSWR) from 2018 to 2022 at 0.1° × 0.1° resolution. The SSWR, indicated by the black rectangle (latitude 5–10° N and longitude 28–34.5° E), encompasses the area of interest.
Remotesensing 16 04744 g002
Remotesensing 16 04744 g003
Figure 3. Monthly CH4 emissions for the South Sudan Wetlands Region (SSWR) derived from TROPOMI data from 2018 to 2022. The Y-axis represents CH4 emissions anomalies in Tg CH4. The anomalies in CH4 emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.
Figure 3. Monthly CH4 emissions for the South Sudan Wetlands Region (SSWR) derived from TROPOMI data from 2018 to 2022. The Y-axis represents CH4 emissions anomalies in Tg CH4. The anomalies in CH4 emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.
Remotesensing 16 04744 g003
Remotesensing 16 04744 g004
Figure 4. The TROPOMI-inferred CH4 enhancements (black), river altimetry (pink), GRACE equivalent water thickness (green), and ENSO index (blue and red) for the South Sudan Wetland Region (SSWR) highlighting the interconnected dynamics. The anomalies in CH4 emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.
Figure 4. The TROPOMI-inferred CH4 enhancements (black), river altimetry (pink), GRACE equivalent water thickness (green), and ENSO index (blue and red) for the South Sudan Wetland Region (SSWR) highlighting the interconnected dynamics. The anomalies in CH4 emissions are calculated as the deviation from the multi-year monthly mean, standardized by dividing the difference by the standard deviation (SD) of the observed values for that particular month across all years. Positive values indicate higher-than-average emissions, while negative values represent lower-than-average emissions.
Remotesensing 16 04744 g004
Remotesensing 16 04744 g005
Figure 5. (a) South Sudan Wetlands Region (SSWR) catchments, and (b) TROPOMI CH4 emission in comparison to river altimetry measurements per catchment.
Figure 5. (a) South Sudan Wetlands Region (SSWR) catchments, and (b) TROPOMI CH4 emission in comparison to river altimetry measurements per catchment.
Remotesensing 16 04744 g005
Remotesensing 16 04744 g006
Figure 6. Correlation between PCR-GLOBWB sub-surface parameters anomalies and TROPOMI CH₄ emission anomalies.
Figure 6. Correlation between PCR-GLOBWB sub-surface parameters anomalies and TROPOMI CH₄ emission anomalies.
Remotesensing 16 04744 g006
Remotesensing 16 04744 g007
Figure 7. Correlation between PCR-GLOBWB surface parameters anomalies and TROPOMI CH₄ emission anomalies.
Figure 7. Correlation between PCR-GLOBWB surface parameters anomalies and TROPOMI CH₄ emission anomalies.
Remotesensing 16 04744 g007
Remotesensing 16 04744 g008
Figure 8. Correlation between PCR-GLOBWB river parameters anomalies and TROPOMI CH₄ emission anomalies.
Figure 8. Correlation between PCR-GLOBWB river parameters anomalies and TROPOMI CH₄ emission anomalies.
Remotesensing 16 04744 g008
Remotesensing 16 04744 g009
Figure 9. Comparison of normalized TROPOMI CH₄ emissions from the SSWR (black) alongside normalized water levels of Lake Victoria (green). The bars indicate the phases of the El Niño Southern Oscillation (ENSO), with El Niño events represented in blue and La Niña events in red, highlighting potential correlations between climatic phenomena and CH₄ emissions.
Figure 9. Comparison of normalized TROPOMI CH₄ emissions from the SSWR (black) alongside normalized water levels of Lake Victoria (green). The bars indicate the phases of the El Niño Southern Oscillation (ENSO), with El Niño events represented in blue and La Niña events in red, highlighting potential correlations between climatic phenomena and CH₄ emissions.
Remotesensing 16 04744 g009
Table 1. PCR-GLOBWB hydrological parameters used in the study.
Table 1. PCR-GLOBWB hydrological parameters used in the study.
PCR-GLOBWB ParameterDescriptionUnit
Sub-surface parametersCapillary riseThe upward movement of water in porous materials due to surface tension.m.month−1
Groundwater RechargeThe process of water percolating through the ground to replenish underground aquifers.m.month−1
Upper soil moistureThe moisture content in the upper soil layer (5 cm).m3.m−3
Lower soil moistureThe moisture content in deeper soil layers (5−30 cm).m3.m−3
Surface water parametersSurface water levelThe water height in rivers, lakes, or other surface water bodies.m
Near-surface soil saturation degreeA measure of how saturated or waterlogged the top layer of soil is expressed as a percentage.%
Flood inundation depthThe depth of water covering an area during a flood event.m
Flood inundation volumeThe total volume of water that inundates an area during a flood.m3
Fraction of surface waterThe proportion of water present on the Earth’s surface in comparison to the total water volume of the cell area.%
River parametersChannel storageThe water within river channels often fluctuates with river flow and precipitation events.m3
DischargeThe volume of water flowing through a particular cross-section in a river or stream.m3.month−1
Table 2. Methane (XCH4) enhancement and emission estimates from 2018 to 2022 from TROPOMI. Data coverage is defined as the fraction of 0.1° × 0.1° grid cells within the SSWR that contain at least five high-quality TROPOMI measurements per quarter. Wind speed data, representing the average boundary layer wind, were obtained from ERA5. Emission estimates were calculated using the methodology described in Pandey et al. (2017), with ± representing 1σ uncertainty.
Table 2. Methane (XCH4) enhancement and emission estimates from 2018 to 2022 from TROPOMI. Data coverage is defined as the fraction of 0.1° × 0.1° grid cells within the SSWR that contain at least five high-quality TROPOMI measurements per quarter. Wind speed data, representing the average boundary layer wind, were obtained from ERA5. Emission estimates were calculated using the methodology described in Pandey et al. (2017), with ± representing 1σ uncertainty.
TimeXCH4 Enhancement (ppb)Coverage (%)Wind Speed (m/s)Emission (Tg/yr)
201817.1 ± 4.355.02.8 ± 0.39.63 ± 2.9
201919.1 ± 4.360.32.7 ± 0.39.29 ± 2.8
202026.7 ± 4.754.62.9 ± 0.217.0 ± 3.5
202122.6 ± 5.259.62.8 ± 0.212.2 ± 3.6
202223.0 ± 5.252.93.0 ± 0.315.6 ± 3.2
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

Albuhaisi, Y.A.Y.; van der Velde, Y.; Pandey, S.; Houweling, S. Integrating Satellite Observations and Hydrological Models to Unravel Large TROPOMI Methane Emissions in South Sudan Wetlands. Remote Sens. 2024, 16, 4744. https://doi.org/10.3390/rs16244744

AMA Style

Albuhaisi YAY, van der Velde Y, Pandey S, Houweling S. Integrating Satellite Observations and Hydrological Models to Unravel Large TROPOMI Methane Emissions in South Sudan Wetlands. Remote Sensing. 2024; 16(24):4744. https://doi.org/10.3390/rs16244744

Chicago/Turabian Style

Albuhaisi, Yousef A. Y., Ype van der Velde, Sudhanshu Pandey, and Sander Houweling. 2024. "Integrating Satellite Observations and Hydrological Models to Unravel Large TROPOMI Methane Emissions in South Sudan Wetlands" Remote Sensing 16, no. 24: 4744. https://doi.org/10.3390/rs16244744

APA Style

Albuhaisi, Y. A. Y., van der Velde, Y., Pandey, S., & Houweling, S. (2024). Integrating Satellite Observations and Hydrological Models to Unravel Large TROPOMI Methane Emissions in South Sudan Wetlands. Remote Sensing, 16(24), 4744. https://doi.org/10.3390/rs16244744

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