Terrain Research Articles

Synoptic Scale Controls and Aerosol Effects on Fog and Low Stratus Life Cycle Processes in the Po Valley, Italy

AbstractFog and low stratus clouds (FLS) form as a result of complex interactions of multiple factors in the atmosphere and at the land surface and impact both the anthropogenic and natural environments. Here, we analyze the role of synoptic conditions and aerosol loading on FLS occurrence and persistence in the Po valley in northern Italy. By applying k‐means clustering to reanalysis data, we find that FLS formation in the Po valley is either based on radiative processes or moisture advection from the Mediterranean sea. Satellite‐based data on FLS persistence shows longer persistence of radiatively formed FLS events, likely due to air mass stagnation and a temperature inversion. Ground‐based aerosol optical depth observations further reveal that FLS event duration is significantly higher under high aerosol loading. The results underline the combined effect of topography, moisture advection and aerosol loading on the FLS life cycle in the Po valley.

Ecological effects of land use and land cover change in the typical ecological functional zones of Egypt

Evaluation the value of ecosystem services can provide crucial reference for formulating ecological protection plans and making informed management decision. This study utilized Landsat remote sensing images as the data source and employed land-use transfer chains and land use transfer matrix methods to compare the ecological effects of land use and land cover change(LUCC) in the two typical ecological functional zones of Egypt, one area isthe northern region of Egypt with the Priority function of Economic Development, another area is the Nasser Reservoir area with the priority function of water conservation. The key findings are as follows: (1)The northern region of Egypt and the Nasser Reservoir exhibited different trends across the ecological functional zones.From 1985 to 2020, the northern region of Egypt is characterized by an increase in arable land and impervious surface,the Nasser Reservoir area is distinguished by the expansion of water and a reduction in sand.(2) Key land transfer patterns vary significantly between the ecological functional zones of the northern region of Egypt and the Nasser Reservoir. From 1985 to 2020, the Egypt’s northern region experienced significant shifts between arable land, sand, and impervious surfaces. The conversion of arable land to impervious surfaces represented 2.41% of the total arable land transfer area, while sand converted to impervious surfaces accounted for 11.68% of the total sand transfer area. In the Nasser Reservoir region, significant conversions occurred between water, grasslands, and sand. Sand converted to water made up 3.83% of the total sand transfer area, while grassland converted to water compriseed 5.38% of the total grassland transfer area. (3) The ecological benefits of the northern region of Egypt and the Nasser Reservoir have increased significantly. In northern region of Egypt, the primary drivers of this increase in ecological service value are the conversion of arable land and grasslands, along with the ecological development and utilization of sand. In the Nasser Reservoir area,the inflow of water and the continued outflow of sands are the main factors contributing to the rise in ecosystem service value. (4) Differentiated LUCC management should be implemented for different ecological functional zones. In northern region of Egypt, the focus should be on balancing ecological security with human development, while the Nasser Reservoir area should prioritize water resource protection and sustainable development. The research findings are not only significant for optimizing LUCC but also for enhancing ecosystem service functions in Egypt.

Characterizing heatwaves based on land surface energy budget

Heat extremes pose pronounced threats to social-ecological systems and are projected to become more intense, frequent, and longer. However, the mechanisms driving heatwaves vary across heatwave types and are not yet fully understood. Here we decompose perturbations in the surface energy budget to categorize global heatwave-days into four distinct types: sunny–humid (38%), sunny-dry (26%), advective (18%), and adiabatic (18%). Notably, sunny-dry heatwave-days decrease net ecosystem carbon uptake by 0.09 gC m−2 day−1 over harvested areas, while advective heatwave-days increase the thermal stress index by 6.20 K in populated regions. In addition, from 2000 to 2020, sunny-dry heatwaves have shown the most widespread increase compared to 1979 to 1999, with 67% of terrestrial areas experiencing a doubling in their occurrence. Our findings highlight the importance of classifying heatwave-days based on their underlying mechanisms, as this can enhance our understanding of heatwaves and improve strategies for heat adaptation.

A comparative study of the effects of urban morphology on land surface temperature in Chengdu and Chongqing, China

Urbanization combined with global climate change, exacerbates the urban thermal environment and hinders sustainable urban development. However, the complex relationships between land surface temperature (LST) and urban morphology are being further understood, particularly in relation to different urban development patterns, distinct topography, and 3D building morphology. Thus, this study conducted a comparative study in Chengdu and Chongqing, Southwest China. We explored the impact of comprehensive factors (including socio-economic factos, topography, land use composition, and building morphology) on LST by employing the methods of linear regression, geographical detector model, and the boosted regression trees. Our results suggest that (1) high LST was mainly observed in the central part of Chengdu but it presented multicenter aggregation trend in Chongqing; (2) Socio-economic factors were the dominant variables affecting LST in both cities; (3) land use composition and building morphology showed distinct contributions to LST among the two cities; and (4) 3D building management was more effective in Chengdu than in Chongqing. A better understanding of the impact of various influencing factors on LST will enable policy makers and planners to develop appropriate strategies for constructing climate-adaptive cities and mitigating urban heat.

Spatial downscaling of SMAP soil moisture to high resolution using machine learning over China’s Loess Plateau

Soil moisture (SM) is a critical physical parameter in land surface processes that affects the atmospheric and hydrological cycles. Soil Moisture Active Passive (SMAP) mission produces high-quality global SM data, but its low spatial resolution limits the applications at a regional or local scale. In this study, we developed a novel method to select optimal factors from 34 candidate downscaling factors for each month using three machine learning methods (Back Propagation Neural Network (BPNN), Support Vector Machine (SVM) and Random Forest (RF)) to produce monthly time series SM products at a spatial resolution of 1 km for the entire Loess Plateau in China from 2015 to 2023. 11 in-situ SM measurements distributed in Loess Plateau and precipitation data were used to evaluate the downscaling performances of the three machine learning approaches. The results show that the spatial trends of the three downscaled SM were essentially the same as the SMAP SM, with similar spatial patterns, and all three downscaled SM can provide more spatial information and texture features than SMAP SM. Among the three types of downscaled SM, the RF downscaled SM reached the highest R (0.654) and the smallest unbiased Root Mean Square Error (ubRMSE) (0.044 m3m−3), better than the BPNN and SVM downscaled SM, and most closely matched the in-situ measurements. The dynamics of SM were successfully captured by RF downscaled SM, which exhibited strong temporal consistency with the in-situ SM and responded well to precipitation events, with significant increases in SM values in the month of high precipitation and subsequent months. In conclusion, the RF model has the best downscaling effect and it can be used to provide high spatial resolution SM data for applications at a regional or local scale across the Loess Plateau.

Observed response of microwave land surface emissivity to antecedent rainfall in Hainan Island

ABSTRACT Microwave emissivity is an important parameter for over-land retrieval of atmospheric variables such as water vapour, rainfall, and snowfall. However, it remains unclear for the dynamic response of multi-frequency microwave emissivity to prior rainfall over heterogeneous land surfaces. This study combined multi-frequency Microwave Land Surface Emissivity (MLSE) under all-weather conditions with hourly in-situ rainfall to investigate the response of MLSE to rainfall over different vegetation types in Hainan Island in China. Specifically, we explored the change of MLSE at 6.925, 10.65, 18.7, 23.8 and 36.5 Ghz to rainfall intensity in four rainfall cases, followed with statistical characteristics of MLSE to rainfall and dry duration (DD) across different vegetation types during 2003–2010. We found that the magnitude of decline in MLSE (0.004–0.06) is dependent on the rainfall intensity and duration, and the reduction caused by short-lasting heavy rain (daily rainfall >100 mm) was 2–6 times higher than that by light long-lasting rain (daily rainfall <80 mm). Among different microwave frequencies, the MLSE at 23.8 Ghz was most reduced by rainfall, while less reduction was found at low microwave frequencies such as at 6.925 and 10.65 Ghz. More importantly, the change of MLSE after rainfall was significantly different among vegetation types. Over woody savannas and forested lands, MLSE slightly decreased (about 0.004) in the first two hours of DD, whereas a rapid increase was observed in the first hour of DD over areas dominated with woody savanna, grassland, and cropland. These findings improve our understanding of the multi-frequency MLSE in response to rainfall over complex land surfaces, and benefit the microwave retrieval of rainfall over land.

Spatial–Temporal Evolution and Driving Force Analysis of Blue–Green Space in the Chengdu–Chongqing Economic Circle, China

In the rapid process of urbanization, revealing the patterns and driving forces behind the evolution of blue–green spaces holds significant value for optimizing urban blue–green environments. This study systematically investigates the spatial–temporal evolution characteristics and driving forces of blue–green space in the Chengdu–Chongqing Economic Circle from 1990 to 2020, utilizing GIS technology, landscape pattern analysis, and geographic detectors. The research findings indicate the following: (1) The area of blue–green space in the study area exhibits a general trend of initial growth followed by decline, with significant changes occurring between 2010 and 2020. (2) The fragmentation degree of blue–green space is gradually increasing, while connectivity among landscapes is decreasing; however, there has been an increase in landscape distribution uniformity. More than 90% of blue–green spaces expanded mainly through adjacency patterns. (3) In examining driving forces, it was found that temperature, topographic relief, elevation, population density, and construction intensity are the primary driving factors. Notably, the influence of natural factors has diminished over time while human social factors have significantly intensified. This study offers solutions for optimizing the configuration of blue–green spaces within the Chengdu–Chongqing Economic Circle. It also serves as a reference case for promoting high-quality urbanization in developing countries undergoing rapid urbanization.

Combined effects of urbanization and climate variability on water and carbon balances in a rice paddy-dominated basin in Southern China

Abstract Urbanization is known to elevate storm runoff, but how it influences carbon cycle and ecosystem productivity through altering the evapotranspiration (ET) process is less clear. We examined the combined effects of urbanization including change in impervious surface area (ISA) and climate variability on the water and carbon balances of the Qinhuai River Basin (QRB) over 2001-2018. QRB represents a typical rice paddy-dominated region that experienced rapid urbanization in southern China. We improved a monthly scale Water Supply Stress Index (WaSSI) ecosystem service model by integrating local eddy flux measurements and high-resolution remote sensing data. We found a significant downward trend in both ET (−4.6 mm yr-1, p&lt;0.05) and gross primary productivity (GPP) (−10.4 gC m-2 yr-1, p&lt;0.05) but a significant upward trend in water yield (Q) (+28.6 mm yr-1, p&lt;0.05). These ecosystem function changes coincided with a 96% increase in urban areas, 23.2% increase in ISA, and a 37% reduction in rice paddy fields. The mean annual watershed GPP decreased from 1048 gC m-2 to 998 gC m-2 while the annual Q increased from 284 mm to 669 mm from 2001 to 2018. Scenario modelling experiments suggested that the negative impacts of loss of rice paddy fields and increase in ISA on ET and GPP overwhelmed the positive impacts of climate warming. The reduction in GPP and increase in Q were largely attributed to the increases in ISA, not necessarily due to changes in land use types (e.g., urban area). The expansion of urban area, increase in ISA and reduction in leaf area index, and increase in precipitation explained the increase in Q. Our research offers insight about the interactions of carbon and water cycles through the critical ET processes under a changing climate and land surface characteristics at a watershed level. Our modelling tool and analysis provides land managers and policy makers information for designing effective ‘Urban Nature-based Solutions’ to mitigate the negative environmental effects of urbanization on carbon and water.

Spatial Assessment of Urban Heat Island (UHI) in Kütahya using Landsat-8 Satellite Data

The Urban Heat Island (UHI) effect refers to the phenomenon where urban areas experience significantly higher temperatures than their rural surroundings due to human activities and modifications, such as replacing natural land cover with impervious surfaces, reducing vegetation, and increasing heat-absorbing materials. Kütahya central district has a dense population and building layout, which contributes to the intense UHI. Geographic Information System (GIS) and its tools are widely used in assessing UHI in urban areas, providing insights for both researchers and local authorities. Landsat-8 images acquired on August 8, 2023 (Path: 179, Row: 33, Cloud Cover: 0%) were utilized to create the Urban Heat Island (UHI) map for the study area. Through spatial analysis using Landsat-8 data for Kütahya city center, six different classes were defined, with the UHI map created using land surface temperature values. The highest and lowest UHI values vary between 4.245°C and -3.457°C. It has been observed that the average UHI increases by 12% in areas lacking dense building structures, characterized by rock surfaces and limited green areas. Conversely, in areas with parks, forests, and sparse settlements, the UHI decreases by 10%. With this study, a comprehensive evaluation of UHI was conducted for Kütahya city center. While the results obtained are expected to be an important resource for researchers and local authorities, they are also anticipated to be beneficial in the fields of engineering geology, urban geology, and urban planning.

Exploring Topography Downscaling Methods for Hyper‐Resolution Land Surface Modeling

AbstractHyper‐resolution land surface modeling provides an unprecedented opportunity to simulate locally relevant water and energy cycles. However, the available meteorological forcing data is often insufficient to fulfill the requirements of hyper‐resolution modeling. Here, we developed a comprehensive downscaling framework based on topography‐adjusted methods and automated machine learning (AutoML). With this framework, a 90 m and hourly atmospheric forcing data set was developed from ERA5 data at a 0.25° resolution, and the Common Land Model (CoLM) was then forced with the developed forcing data over two complex terrain regions (the Heihe River Basin and Upper Colorado River Basin). We systematically evaluated the downscaled forcing and the CoLM outputs against both in situ observations and gridded data. The ground‐based validation results suggested consistent improvements for all downscaled forcing variables with mean RMSE improved by 6.362%–95.86%. The downscaled forcings, which incorporated detailed topographic features, offered improved magnitude estimates, achieving a comparable level of performance to that of regional reanalysis forcing data. The downscaled forcing driving the CoLM model showed comparable or better skills in simulating water and energy fluxes, as verified by in situ validations. The hyper‐resolution simulations provided a detailed and more reasonable description of land surface processes and attained similar spatial patterns and magnitudes with high‐resolution land surface data, especially over highly elevated areas. Additionally, this study highlighted the benefits of using mountain radiation theory‐based shortwave radiation downscaling models and AutoML‐assisted precipitation downscaling models. These findings emphasized the significance of integrating topography‐based downscaling methods for hillslope‐scale simulations.

Identification of dominant drivers of streamflow spatiotemporal variations in typical mountainous areas in the Hexi Corridor, China

Study regionTypical mountain areas in the Hexi Corridor, China. Study focusWater security and ecosystem sustainability of arid inland river basins are highly dependent on upstream streamflow. However, due to the complex geographical environment and limited observation data in the study region, the attribution of spatiotemporal variations in streamflow influenced by climate change and/or human activities remains unclear. Here, we used partial least squares regression (PLSR) and the Budyko framework to unravel the dominant drivers of spatiotemporal variation in streamflow over the past 30 yr. New hydrological insight for the regionPrecipitation, topographic wetness index, slope, forest land, gross primary productivity, hydrological connectivity, soil organic carbon content, silt content, relative relief, NDVI and gravel content dominated spatial variation in streamflow. Temporal variation of streamflow was sensitive to precipitation and land surface. Specifically, increased precipitation and land surface alteration dominated the increase in streamflow in 50 % of the watersheds and the decrease in streamflow in 33 % of them, respectively. Further, land surface alteration was dominated by expansion of agricultural and built-up areas, weakened hydrological connectivity, increased landscape aggregation and forest cover. Controlling agricultural and built-up areas and the scale of afforestation, and focusing on the dynamics of hydrological connectivity and landscape patterns in the upstream reaches are imperative to maintain the security and sustainability of water resources in the arid inland river basins.

Exploring El Nino effects on agricultural area using Landsat images analysis: A case study in Bondowoso Regency, Indonesia

El Nino, which hit Indonesia in 2023, poses severe food security threats due to the high dry season with no rainfall and minimal cloud cover and can trigger serious drought problems if it happen for a long time. This study aimed to explore the impact on agricultural land in Bondowoso Regency during El Nino events. The analysis in this study primarily uses land surface temperature (LST) and normalized difference vegetative index (NDVI) map distribution. The Landsat data from USGS are collected and processed to become LST and NDVI distribution maps. Data analysis focused on the agricultural area layers based on data from the Indonesia geospatial portal. Referring to the LST and NDVI map distribution, the notable rise of LST starts in August 2023, and the peak is in October 2023. Around 46% of areas in the Bondowoso regency are detected as hotspot areas, which had LST above 30&lt;sup&gt;o&lt;/sup&gt;C in October 2023. El Nino affects the irrigated lands and rain-fed fields more than the plantations. The NDVI alteration data does not show that the Bondowoso Regency is experiencing extraordinary drought due to the short-term impact of El Nino. However, the emergence of numerous areas in the moderate NDVI category warns that stress affecting vegetation is starting to occur. Mitigation plans should be prepared for the long-term impact of El Nino, particularly in the hotspot areas. This study could be a comprehension tool for the government and farmers to prepare mitigation plans.

Toward an advanced physics-based scheme for retrieving land surface emissivity and temperature based on FengYun-3D MERSI-II daytime mid-infrared data

Toward an advanced physics-based scheme for retrieving land surface emissivity and temperature based on FengYun-3D MERSI-II daytime mid-infrared data

Hysteresis in seasonal land-atmospheric interactions over India and its characteristics across croplands and forests

Abstract Satellite-derived Vegetation Optical Depth and Soil Moisture (SM) data reveal the critical role of the soil-vegetation continuum in storing rainwater during the Indian Summer Monsoon and supporting evapotranspiration (ET) during the dry non-monsoon season. During the non-monsoon drier period, the climatologically estimated spatial mean of ET exceeds precipitation input, a phenomenon known as the Soil-Vegetation Capacitor (SVC) effect, which is pivotal in maintaining ecosystem productivity. Notably, our analysis reveals significant variations in the capacitor period between croplands and forests, with croplands exhibiting a ~77 days longer due to dual crop seasons influenced by regional precipitation. The well-recognized hysteresis curves, observed in magnetization and soil-water characteristic curves (SWCC), highlight phenomena where a system's state is influenced by its historical inputs or states and are integral to our findings. We report a previously undocumented seasonal hysteresis in the relationship between the evaporative fraction (EVF) and SM for Indian croplands and forests. We further found that the croplands SM-EVF relation exhibits a reversal in hysteresis in the case of root-zone SM. The surface SM-EVF hysteresis is not present in forests with large root depths and reduced soil evaporation due to high canopy shading, and yet it is present for the root-zone SM. With its reversal for croplands, the newly found hysteresis must be addressed in redefining the critical SM threshold to demarcate the energy and water-limiting regimes. It should be incorporated in the land surface modeling parameterization. Additionally, we observed hysteresis in the soil moisture-gross primary productivity (SM-GPP) relationship across both land covers and soil profiles (surface and root-zone), underscoring the need to investigate such processes to consider their dynamics in future ecological and hydrological models.

Gap-filling of land surface temperature in arid regions by combining Landsat 8 and 9 imageries

Abstract Land surface temperature (LST) is an important factor in land monitoring studies, but due to the presence of clouds, dust and sensor issues, there are missing values. The aims of this research are to determine the optimal parameters for the reconstruction of Landsat-LST images, required in many applications, by the harmonic analysis of time series algorithm (HANTS) and to investigate the possibility of improving LST reconstruction accuracy using Landsat 8 and 9 images simultaneously. For these aims, 91 Landsat 8 and 9 images with 100 m spatial resolution in 2022 and 2023 are employed, covering Yazd-Ardakan plain in Iran. Three methods are used for evaluation. In method one, a part of LST image is considered as a gap and is compared with the initial value after reconstruction. In method two, on a cloudy day and a cloudless day, surface temperature values are measured using thermometers at fifty points in plain lands, and the difference between gap-filled satellite measurements and ground measurements is calculated. In method three, all the reconstructed LST images are compared with the original images. In method one, the root mean square error (RMSE) of reconstructed LST reduces by 1.3°C when using the combined Landsat 8 and 9 images. In method two, RMSEs of reconstructed LST images are 6.1°C when using Landsat 8 and 5.4°C when using the combined Landsat 8 and 9. Method three shows that 41% of the study region has RMSE of less than 2°C when using only Landsat 8, while this value becomes 72% when combining Landsat 8 and 9. In general, the combined use of Landsat 8 and 9 LST images improves the accuracy of reconstruction using HANTS. The findings of this research are crucial for regional applications and remote monitoring of surface temperature in areas with limited weather stations.

Assessing long-term water storage dynamics in Afghanistan: An integrated approach using machine learning, hydrological models, and remote sensing

Assessment of Terrestrial Water Storage (TWS) components is crucial for understanding regional climate and water resources, particularly in arid and semi-arid regions like Afghanistan. Given the scarcity of ground-based data, this study leverages remote sensing datasets to quantify water storage changes. We integrated Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-on (GRACE-FO) data with WaterGap, Global Land Water Storage (GWLS), Catchment Land Surface Model (CLSM), and climate variables (precipitation, temperature, potential evapotranspiration) using artificial neural networks (ANN) and random forests (RF). Additionally, Ice, Cloud, and Land Elevation Satellite (ICESat-1,2) data were utilized to estimate glacier mass changes. Seasonal trend decomposition using Loess (STL) was applied to assess TWS changes from 2003 to 2022. Our methodology reveals a high correlation (R = 0.90–0.97) between reconstructed and observed TWS anomalies across Afghanistan's major basins. Glacier mass decreased by −0.59 and −1.17 Gt/year during 2003–2009 and 2018–2022, respectively, while overall TWS declined by −2.46 Gt/year. The HRB experienced the largest TWS loss (−1.47 Gt/year), primarily due to groundwater depletion (−1.18 Gt/year). These findings underscore the importance of our approach for assessing water resources, providing vital insights for sustainable management under a changing climate in a data-scarce country.

A Modified Method for Reducing the Scale Effect in Land Surface Temperature Downscaling at 10 m Resolution

Satellite-derived Land Surface Temperature (LST) plays an important role in research on natural energy balance and water cycle. Considering the tradeoff between spatial and temporal resolutions, accurate fine-resolution LST must be obtained through the use of LST downscaling (DLST) technology. Various methods have been proposed for DLST at fine resolutions (e.g., 10 m) and small scales. However, the scale effect of these methods, which is inherent to DLST processes at different extents, has rarely been addressed, thus limiting their application. In this study, a modified daily 10 m resolution DLST method based on Google Earth Engine, called mDTSG, is proposed in order to reduce the scale effect at fine spatial resolutions. The proposed method introduces a convolution-based moving window into the DLST process for the fusion of different remote sensing data. The performance of the modified method is compared with the original method in six regions characterized by various extents and landscape heterogeneity. The results show that the scale effect is significant in the DLST process at fine resolutions across extents ranging from 100 km2 to 22,500 km2. Compared with the original method, mDTSG can effectively reduce the LST value differences between tile edges, especially when considering large extents (&gt;22,500 km2) with an average R2 improvement of 33.75%. The average MAE is 1.63 °C, and the average RMSE is 2.3 °C in the mDTSG results, when compared with independent remote sensing products across the six regions. A comparison with in situ observations also shows promising results, with an MAE of 2.03 °C and an RMSE of 2.63 °C. These findings highlight the robustness and scalability of the mDTSG method, making it a valuable tool for fine-resolution LST applications in diverse and extensive landscapes.

Response of Upwelling Parameter Before, During, and After Tropical Cyclone (Case Study: Tropical Cyclone Marcus)

Tropical cyclones (TC) can cause damage when they are on the land surface, but on the other hand TC contribute to ocean productivity through upwelling or downwelling when they pass through the ocean. Information of the location and timing of upwelling and downwelling is important for fishing - activities estimation. High chlorophyll-a concentrations and low sea surface temperatures are proxies for upwelling events in the ocean. This study aims to investigate the effects of the TC Marcus on the chlorophyll-a distribution in the Timor Sea using a combination of remotely sensed data from Aqua MODIS chlorophyll-a and ocean-atmosphere reanalysis outputs. The results showed that wind speed, sea surface temperature, salinity, and ocean currents were increased during the TC Marcus event. Spatial analysis reveals that the concentration of chlorophyll-a is high in the waters of the Timor Sea, at coordinates around 12°–13° S and 125°–129° E. High chlorophyll-a concentrations occurred before and after TC Marcus event, according to temporal analysis. The distribution of chlorophyll-a concentrations decreased on March 17–24, 2018, during the occurrence of the TC Marcus in the Timor Sea.

From Data to Decision: Interpretable Machine Learning for Predicting Flood Susceptibility in Gdańsk, Poland

Flood susceptibility prediction is complex due to the multifaceted interactions among hydrological, meteorological, and urbanisation factors, further exacerbated by climate change. This study addresses these complexities by investigating flood susceptibility in rapidly urbanising regions prone to extreme weather events, focusing on Gdańsk, Poland. Three popular ML techniques, Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Networks (ANN), were evaluated for handling complex, nonlinear data using a dataset of 265 urban flood episodes. An ensemble filter feature selection (EFFS) approach was introduced to overcome the single-method feature selection limitations, optimising the selection of factors contributing to flood susceptibility. Additionally, the study incorporates explainable artificial intelligence (XAI), namely, the Shapley Additive exPlanations (SHAP) model, to enhance the transparency and interpretability of the modelling results. The models’ performance was evaluated using various statistical measures on a testing dataset. The ANN model demonstrated a superior performance, outperforming the RF and the SVM. SHAP analysis identified rainwater collectors, land surface temperature (LST), digital elevation model (DEM), soil, river buffers, and normalized difference vegetation index (NDVI) as contributors to flood susceptibility, making them more understandable and actionable for stakeholders. The findings highlight the need for tailored flood management strategies, offering a novel approach to urban flood forecasting that emphasises predictive power and model explainability.

Quantifying Annual Glacier Mass Change and Its Influence on the Runoff of the Tuotuo River

Glacier meltwater is an indispensable water supply for billions of people living in the catchments of major Asian rivers. However, the role of glaciers on river runoff regulation is seldom investigated due to the lack of annual glacier mass balance observation. In this study, we employed an albedo-based model with a daily land surface albedo dataset to derive the annual glacier mass balance over the Tuotuo River Basin (TRB). During 2000–2022, an annual glacier mass balance range of −0.89 ± 0.08 to 0.11 ± 0.11 m w.e. was estimated. By comparing with river runoff records from the hydrometric station, the contribution of glacier mass change to river runoff was calculated to be 0.00–31.14% for the studied period, with a mean value of 9.97%. Moreover, we found that the mean contribution in drought years is 20.07%, which is approximately five times that in wet years (4.30%) and twice that in average years (9.49%). Therefore, our results verify that mountain glaciers act as a significant buffer against drought in the TRB, at least during the 2000–2022 period.