Nehar Mandal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nehar Mandal has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Kironmala Chanda
Department of Civil Engineering, Indian Institute of Technology (Indian School of Mines), Dhanbad, India
5 shared publications - Prabal Das
Department of Civil Engineering The University of Texas, Arlington, TX, USA, 76019
5 shared publications Kironmala Chanda
3 shared publications- Kironmala Chanda
Department of Civil Engineering, Indian Institute of Technology (Indian School of Mines) Dhanbad, India
1 shared publication - Nirup Sundar Mandal1 shared publication
- Sachidanand Kumar
Department of Civil Engineering, Indian Institute of Technology (Indian School of Mines), Dhanbad, India
1 shared publication
Latest Publications
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Evolution of Heat wave characteristics across South Asia and identification of the most affected cities in the recent decade
Heat wave (HW) is a hazardous climate extreme that can lead to serious impacts on human health, posing challenges to the UN Sustainable Development Goals #3, #11, and #13. This study examines the heat wave characteristics across South Asia and surrounding regions during a 45-year period (1980-2024) with a particular focus on recent intensification and increasing population exposure. Daily 2-m air temperature data of European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA5) was used for identification of hot days and resultant heat waves. The annual count of hot days (DH) is the number of days the daily maximum temperature surpasses 90th percentile daily maximum temperature, whereas hot nights (NH) refers to exceedance of 90th percentile daily minimum temperature, calculated over a 15-day moving window representing long term climatology of the time of the year being considered. A minimum of three consecutive compound hot day and nights was identified as a HW event. Three HW indices were computed annually: the number of HW events (HWn), the number of HW participating days (HWp), and HW magnitude (HWm), which accounts for combined daytime and nighttime temperature departures. For each grid location, these indices were aggregated at decadal scales from the 1980s to the 2020s to examine the evolution of the HW characteristics across the study domain.The study revealed that globally, NH has increased substantially (266.19%) from 1980s to 2020s leading to more frequent HWs (44,907 events/year in 1980s to 333,424 events/year in 2020s) during the study period. In Peninsular India, HWn was found to be as high as 98 events and HWp was as high as 533 days in the 2020s. HWm was even more than 500 °C² in some locations in Eastern Asia during the same decade, indicating that both day-time and night-time temperatures showed large anomalies with respect to the long-term climatology.To quantify the impact of rising heatwaves on rising population, gridded population data from WorldPop of University of Southampton was used to determine the change in population exposure to the HW indices over the recent decade (2015–2024). The major cities with marked increase in population exposure to HW occurrences (i.e., HWn) were identified as Zhengzhou (China), Chengdu (China), Dhaka (Bangladesh), Faridabad (India) and Lahore (Pakistan) with exposure changes ranging from 2.92 × 107 person-events to 6.34 × 107 person-events. The maximum change in population exposure to DH is in Istanbul, Turkey (1.57 × 108 person-days) whereas the same for NH is in Ho Chi Minh City, Vietnam (5.61 × 108 person-days). The rising exposure to NH indicates that many cities are losing the ability of natural night-time cooling and require targeted intervention. Thus, this study offers valuable insights on the spatial and temporal evolution of heat wave characteristics across the most densely populated regions of the world and is expected to be useful for developing policies on climate-resilient urban infrastructure planning.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessEvolution of Heat wave characteristics across South Asia and identification of the most affected cities in the recent decade
Heat wave (HW) is a hazardous climate extreme that can lead to serious impacts on human health, posing challenges to the UN Sustainable Development Goals #3, #11, and #13. This study examines the heat wave characteristics across South Asia and surrounding regions during a 45-year period (1980-2024) with a particular focus on recent intensification and increasing population exposure. Daily 2-m air temperature data of European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA5) was used for identification of hot days and resultant heat waves. The annual count of hot days (DH) is the number of days the daily maximum temperature surpasses 90th percentile daily maximum temperature, whereas hot nights (NH) refers to exceedance of 90th percentile daily minimum temperature, calculated over a 15-day moving window representing long term climatology of the time of the year being considered. A minimum of three consecutive compound hot day and nights was identified as a HW event. Three HW indices were computed annually: the number of HW events (HWn), the number of HW participating days (HWp), and HW magnitude (HWm), which accounts for combined daytime and nighttime temperature departures. For each grid location, these indices were aggregated at decadal scales from the 1980s to the 2020s to examine the evolution of the HW characteristics across the study domain.The study revealed that globally, NH has increased substantially (266.19%) from 1980s to 2020s leading to more frequent HWs (44,907 events/year in 1980s to 333,424 events/year in 2020s) during the study period. In Peninsular India, HWn was found to be as high as 98 events and HWp was as high as 533 days in the 2020s. HWm was even more than 500 °C² in some locations in Eastern Asia during the same decade, indicating that both day-time and night-time temperatures showed large anomalies with respect to the long-term climatology.To quantify the impact of rising heatwaves on rising population, gridded population data from WorldPop of University of Southampton was used to determine the change in population exposure to the HW indices over the recent decade (2015–2024). The major cities with marked increase in population exposure to HW occurrences (i.e., HWn) were identified as Zhengzhou (China), Chengdu (China), Dhaka (Bangladesh), Faridabad (India) and Lahore (Pakistan) with exposure changes ranging from 2.92 × 107 person-events to 6.34 × 107 person-events. The maximum change in population exposure to DH is in Istanbul, Turkey (1.57 × 108 person-days) whereas the same for NH is in Ho Chi Minh City, Vietnam (5.61 × 108 person-days). The rising exposure to NH indicates that many cities are losing the ability of natural night-time cooling and require targeted intervention. Thus, this study offers valuable insights on the spatial and temporal evolution of heat wave characteristics across the most densely populated regions of the world and is expected to be useful for developing policies on climate-resilient urban infrastructure planning.
- 7 February 2026
- Published by Springer Nature in Soft Computing
- Vol. 30 (4) , 2701-2719
Empirical and ML models for estimation of evapotranspiration: comparison of performance and evaluation of spatial transferability
- 7 February 2026
- Published by Springer Nature in Soft Computing
- Vol. 30 (4) , 2701-2719
Empirical and ML models for estimation of evapotranspiration: comparison of performance and evaluation of spatial transferability
- Top Cited PapersOpen Access
- 13 June 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (6) , 2575-2604
Machine-learning-based reconstruction of long-term global terrestrial water storage anomalies from observed, satellite and land-surface model data
Understanding long-term terrestrial water storage (TWS) variations is vital for investigating hydrological extreme events, managing water resources and assessing climate change impacts. However, the limited data duration from the Gravity Recovery and Climate Experiment (GRACE) and its follow-on mission (GRACE-FO) poses challenges for comprehensive long-term analysis. In this study, we reconstruct TWS anomalies (TWSAs) for the period from January 1960 to December 2022, thereby filling data gaps between the GRACE and GRACE-FO missions and generating a complete dataset for the pre-GRACE era. The workflow involves identifying optimal predictors from land surface model (LSM) outputs, meteorological variables and climatic indices using a novel Bayesian network (BN) technique for raster-based TWSA simulations. Climate indices, like the Oceanic Niño Index and Dipole Mode Index, are selected as optimal predictors for a large number of grid cells globally, along with TWSAs from LSM outputs. The most effective machine learning (ML) algorithms among convolutional neural network (CNN), support vector regression (SVR), extra trees regressor (ETR) and stacking ensemble regression (SER) models are evaluated at each grid cell to achieve optimal reproducibility. Globally, ETR performs best for most of the grid cells; this is also noticed at the river basin scale, particularly for the Ganga–Brahmaputra–Meghna, Godavari, Krishna, Limpopo and Nile river basins. The simulated TWSAs (BNML_TWSA) outperformed the TWSAs from LSM outputs when evaluated against GRACE datasets. Improvements are particularly noted in river basins such as the Godavari, Krishna, Danube and Amazon, with median correlation coefficient, Nash–Sutcliffe efficiency, and RMSE values for all grid cells in the Godavari Basin, India, being 0.927, 0.839 and 63.7 mm, respectively. A comparison with TWSAs reconstructed in recent studies indicates that the proposed BNML_TWSA outperforms them globally as well as for all of the 11 major river basins examined. Furthermore, the uncertainty of BNML_TWSA is assessed for each grid cell in terms of the standard error. Results show smaller standard error magnitudes in grid cells in arid regions compared to other regions. The presented gridded dataset is published at https://doi.org/10.6084/m9.figshare.25376695 (Mandal et al., 2024), featuring a spatial resolution of 0.50° × 0.50° and offering global coverage.
- 13 June 2025
- Published by Copernicus GmbH in Earth System Science Data
- Vol. 17 (6) , 2575-2604
Top Cited PapersOpen AccessMachine-learning-based reconstruction of long-term global terrestrial water storage anomalies from observed, satellite and land-surface model data
Understanding long-term terrestrial water storage (TWS) variations is vital for investigating hydrological extreme events, managing water resources and assessing climate change impacts. However, the limited data duration from the Gravity Recovery and Climate Experiment (GRACE) and its follow-on mission (GRACE-FO) poses challenges for comprehensive long-term analysis. In this study, we reconstruct TWS anomalies (TWSAs) for the period from January 1960 to December 2022, thereby filling data gaps between the GRACE and GRACE-FO missions and generating a complete dataset for the pre-GRACE era. The workflow involves identifying optimal predictors from land surface model (LSM) outputs, meteorological variables and climatic indices using a novel Bayesian network (BN) technique for raster-based TWSA simulations. Climate indices, like the Oceanic Niño Index and Dipole Mode Index, are selected as optimal predictors for a large number of grid cells globally, along with TWSAs from LSM outputs. The most effective machine learning (ML) algorithms among convolutional neural network (CNN), support vector regression (SVR), extra trees regressor (ETR) and stacking ensemble regression (SER) models are evaluated at each grid cell to achieve optimal reproducibility. Globally, ETR performs best for most of the grid cells; this is also noticed at the river basin scale, particularly for the Ganga–Brahmaputra–Meghna, Godavari, Krishna, Limpopo and Nile river basins. The simulated TWSAs (BNML_TWSA) outperformed the TWSAs from LSM outputs when evaluated against GRACE datasets. Improvements are particularly noted in river basins such as the Godavari, Krishna, Danube and Amazon, with median correlation coefficient, Nash–Sutcliffe efficiency, and RMSE values for all grid cells in the Godavari Basin, India, being 0.927, 0.839 and 63.7 mm, respectively. A comparison with TWSAs reconstructed in recent studies indicates that the proposed BNML_TWSA outperforms them globally as well as for all of the 11 major river basins examined. Furthermore, the uncertainty of BNML_TWSA is assessed for each grid cell in terms of the standard error. Results show smaller standard error magnitudes in grid cells in arid regions compared to other regions. The presented gridded dataset is published at https://doi.org/10.6084/m9.figshare.25376695 (Mandal et al., 2024), featuring a spatial resolution of 0.50° × 0.50° and offering global coverage.
- Open Access
- 20 January 2025
- Published by Copernicus GmbH
Performance of two-step technique for gap-filling and reconstruction of basin-scale Terrestrial Water Storage Anomalies (TWSA)
Accurate estimation of terrestrial water storage anomalies (TWSA) is essential for the assessment of hydrological extreme events, managing water resources, and evaluating climate change impacts. In this study, a two-step method is applied for the reconstruction of a gridded TWSA product in two study basins: Godavari (GRB), a tropical river basin in India, and the Murray-Darling river basin (MDRB) in Australia. In the first step, the probabilistic dependence structure of the target TWSA and 15 potential predictor variables is developed through Bayesian Network technique to obtain the optimal features which strongly influence the target. In the second step, the potential of Machine Learning (ML) algorithms is utilized to obtain TWSA values, considering the grid-specific features selected in the first step as input. The input set of potential predictors includes monthly TWSA simulations from Global Land Data Assimilation System (GLDAS) Catchment Land Surface Model (i.e., CTWSA) and the GLDAS Noah Land Surface Model (i.e., NTWSA) as well as meteorological variables such as precipitation and temperature for a lead time of up to 2 months and large scale climate indices such as Dipole Mode Index, North Atlantic Oscillation index, and Oceanic Niño Index (ONI). For both study basins, CTWSA and ONI are prominent features selected by the Bayesian Network that influence TWSA. After obtaining the optimal features, Machine Learning (ML) algorithms such as Convolutional Neural Network (CNN), Support Vector Regression (SVR), Extra Trees Regressor (ETR), and Stacking Ensemble Regression (SER) are employed to derive TWSA values (henceforth named as BNML_TWSA). The performances of BNML_TWSA, as well as CTWSA and NTWSA, are evaluated against GRACE TWSA for both study basins using performance metrics such as the Correlation Coefficient (R), Nash–Sutcliffe Efficiency (NSE), and Root Mean Square Error (RMSE). At GRB, ETR demonstrates superior performance at most of the grids (74.3%), followed by SVR (21.1%). In contrast, at MDRB, all four ML algorithms show similar performance: CNN, SVR, ETR, and SER, each being selected as the best models at 25.9%, 21.4%, 26.1%, and 26.6% of the grids respectively. When evaluated against GRACE TWSA, the median values of R for NTWSA, CTWSA, and BNML_TWSA across all grids are 0.78, 0.90, and 0.93, respectively, at the GRB. Similarly, for the MDRB, these values are 0.79, 0.85, and 0.87, respectively. At the GRB, the best NSE value is obtained for BNML_TWSA (0.84), while the lowest performance is observed for NTWSA (0.475). At the MDRB also, the least performance is shown by NTWSA with an RMSE value of 57.3 mm/month, and the best performance is achieved by BNML_TWSA with an RMSE of 35.0 mm/month. The proposed two-step method offers dependable estimates of TWSA compared to land surface models and hydrological models. Hence, the reconstructed TWSA (1960-2022) proves valuable during the data gap period between GRACE and GRACE-FO and the pre-GRACE period.
- 20 January 2025
- Published by Copernicus GmbH
Open AccessPerformance of two-step technique for gap-filling and reconstruction of basin-scale Terrestrial Water Storage Anomalies (TWSA)
Accurate estimation of terrestrial water storage anomalies (TWSA) is essential for the assessment of hydrological extreme events, managing water resources, and evaluating climate change impacts. In this study, a two-step method is applied for the reconstruction of a gridded TWSA product in two study basins: Godavari (GRB), a tropical river basin in India, and the Murray-Darling river basin (MDRB) in Australia. In the first step, the probabilistic dependence structure of the target TWSA and 15 potential predictor variables is developed through Bayesian Network technique to obtain the optimal features which strongly influence the target. In the second step, the potential of Machine Learning (ML) algorithms is utilized to obtain TWSA values, considering the grid-specific features selected in the first step as input. The input set of potential predictors includes monthly TWSA simulations from Global Land Data Assimilation System (GLDAS) Catchment Land Surface Model (i.e., CTWSA) and the GLDAS Noah Land Surface Model (i.e., NTWSA) as well as meteorological variables such as precipitation and temperature for a lead time of up to 2 months and large scale climate indices such as Dipole Mode Index, North Atlantic Oscillation index, and Oceanic Niño Index (ONI). For both study basins, CTWSA and ONI are prominent features selected by the Bayesian Network that influence TWSA. After obtaining the optimal features, Machine Learning (ML) algorithms such as Convolutional Neural Network (CNN), Support Vector Regression (SVR), Extra Trees Regressor (ETR), and Stacking Ensemble Regression (SER) are employed to derive TWSA values (henceforth named as BNML_TWSA). The performances of BNML_TWSA, as well as CTWSA and NTWSA, are evaluated against GRACE TWSA for both study basins using performance metrics such as the Correlation Coefficient (R), Nash–Sutcliffe Efficiency (NSE), and Root Mean Square Error (RMSE). At GRB, ETR demonstrates superior performance at most of the grids (74.3%), followed by SVR (21.1%). In contrast, at MDRB, all four ML algorithms show similar performance: CNN, SVR, ETR, and SER, each being selected as the best models at 25.9%, 21.4%, 26.1%, and 26.6% of the grids respectively. When evaluated against GRACE TWSA, the median values of R for NTWSA, CTWSA, and BNML_TWSA across all grids are 0.78, 0.90, and 0.93, respectively, at the GRB. Similarly, for the MDRB, these values are 0.79, 0.85, and 0.87, respectively. At the GRB, the best NSE value is obtained for BNML_TWSA (0.84), while the lowest performance is observed for NTWSA (0.475). At the MDRB also, the least performance is shown by NTWSA with an RMSE value of 57.3 mm/month, and the best performance is achieved by BNML_TWSA with an RMSE of 35.0 mm/month. The proposed two-step method offers dependable estimates of TWSA compared to land surface models and hydrological models. Hence, the reconstructed TWSA (1960-2022) proves valuable during the data gap period between GRACE and GRACE-FO and the pre-GRACE period.
- 22 May 2024
- Published by Wiley in International Journal of Climatology
- Vol. 44 (9) , 2876-2896
Joint probabilistic behaviour of climate extremes over the Godavari River basin, India
Climate change has an adverse effect on the global hydrological cycle and hence climatic extremes are becoming more persistent. Therefore, the present study analyses the characteristics of the compound extremes in India using the Godavari River basin as a test bed. The study examines the individual return period (IRP) and pairwise joint return period (JRP) of several climate extreme indices such as consecutive dry days (CDD), consecutive wet days (CWD), percentage rainfall of very dry days (R5Tot) and very hot days (TX95) considering historical and future period of two scenarios, SSP245 and SSP585, from the CMIP6 consortium. Both SSPs indicates an upward trend in the indices for CWD and TX95, whilst showcasing a declining trend for CDD and R5Tot at 5% significance level. Under SSP585, the values of JRPs , , and ranges from 3 to 5 years, 14–55 years, 35–274 years and 104 to >1000 during epoch‐1 (2023–2060). During epoch‐2 (2061–2100), the JRPs values are generally higher. Regions displaying lower JRPs values suggest a higher likelihood of simultaneous occurrences of extremes such as floods and droughts, as well as very dry days and warm days within the same year. Under SSP585, the maximum (minimum) value of change rate (CR) of JRP for and during epoch‐1 is more than 900% (−80%) and 1000% (−90%), whereas during epoch‐2, the corresponding CR values are greater than 700% (−80%) and 1000% (−90%). The future possibility of co‐occurrence of ‘CDD and CWD’ and ‘R5Tot and TX95’, both exceeding their 10‐ and 25‐year IRP, in a single year, is highly likely for the regions with low CR values and unlikely for the regions with high CR values. The study would help in understanding the spatial and temporal variability of compound climate extremes over the Godavari River basin.
- 22 May 2024
- Published by Wiley in International Journal of Climatology
- Vol. 44 (9) , 2876-2896
Joint probabilistic behaviour of climate extremes over the Godavari River basin, India
Climate change has an adverse effect on the global hydrological cycle and hence climatic extremes are becoming more persistent. Therefore, the present study analyses the characteristics of the compound extremes in India using the Godavari River basin as a test bed. The study examines the individual return period (IRP) and pairwise joint return period (JRP) of several climate extreme indices such as consecutive dry days (CDD), consecutive wet days (CWD), percentage rainfall of very dry days (R5Tot) and very hot days (TX95) considering historical and future period of two scenarios, SSP245 and SSP585, from the CMIP6 consortium. Both SSPs indicates an upward trend in the indices for CWD and TX95, whilst showcasing a declining trend for CDD and R5Tot at 5% significance level. Under SSP585, the values of JRPs , , and ranges from 3 to 5 years, 14–55 years, 35–274 years and 104 to >1000 during epoch‐1 (2023–2060). During epoch‐2 (2061–2100), the JRPs values are generally higher. Regions displaying lower JRPs values suggest a higher likelihood of simultaneous occurrences of extremes such as floods and droughts, as well as very dry days and warm days within the same year. Under SSP585, the maximum (minimum) value of change rate (CR) of JRP for and during epoch‐1 is more than 900% (−80%) and 1000% (−90%), whereas during epoch‐2, the corresponding CR values are greater than 700% (−80%) and 1000% (−90%). The future possibility of co‐occurrence of ‘CDD and CWD’ and ‘R5Tot and TX95’, both exceeding their 10‐ and 25‐year IRP, in a single year, is highly likely for the regions with low CR values and unlikely for the regions with high CR values. The study would help in understanding the spatial and temporal variability of compound climate extremes over the Godavari River basin.
- Open Access
- 17 May 2024
- Published by Copernicus GmbH in EGUsphere
- Vol. 2024, 1-34
Optimal feature selection for improved ML based reconstruction of Global Terrestrial Water Storage Anomalies
Understanding long-term Terrestrial water storage (TWS) variations is vital for investigating hydrological extreme events, managing water resources, and assessing climate change impacts. However, the limited data duration from the Gravity Recovery and Climate Experiment (GRACE) and its follow-on missions (GRACE-FO) poses challenges for comprehensive long-term analysis. In this study, we reconstruct TWS anomalies (TWSA) for the period Jan 1960 to Dec 2022 thereby filling data gaps between GRACE and GRACE-FO missions as well as generating a complete dataset for the pre-GRACE era. The workflow involves identifying optimal predictors from land surface model (LSM) outputs, meteorological variables, and climatic indices using a novel Bayesian Network (BN) technique for grid-based TWSA simulations. Climate indices, like the Oceanic Niño Index and Dipole Mode Index, are selected as optimal predictors for a large number of grids globally, along with TWSA from LSM outputs. The most effective machine learning (ML) algorithms among Convolutional Neural Network (CNN), Support Vector Regression (SVR), Extra Trees Regressor (ETR), and Stacking Ensemble Regression (SER) models are evaluated at each grid location to achieve optimal reproducibility. Globally, ETR performs best for most of the grids which is also noticed at the river-basin scale, particularly for the Ganga-Brahmaputra-Meghana, Godavari, Krishna, Limpopo, and Nile river basins. The simulated TWSA (BNML_TWSA) outperformed the TWSA from LSM outputs when evaluated against GRACE datasets. Improvements are particularly noted in the river basins such as Godavari, Krishna, Danube, Amazon, etc., with median values of the correlation coefficient, Nash-Sutcliffe efficiency, and RMSE for all grids in Godavari, India, being 0.927, 0.839, and 63.7 mm respectively. A comparison with TWSA reconstructed in recent studies indicates that the proposed BNML_TWSA outperforms them globally as well as for all the 11 major river basins examined. The presented dataset is published at https://doi.org/10.6084/m9.figshare.25376695 (Mandal et al., 2024) and updates will be published when needed.
- 17 May 2024
- Published by Copernicus GmbH in EGUsphere
- Vol. 2024, 1-34
Open AccessOptimal feature selection for improved ML based reconstruction of Global Terrestrial Water Storage Anomalies
Understanding long-term Terrestrial water storage (TWS) variations is vital for investigating hydrological extreme events, managing water resources, and assessing climate change impacts. However, the limited data duration from the Gravity Recovery and Climate Experiment (GRACE) and its follow-on missions (GRACE-FO) poses challenges for comprehensive long-term analysis. In this study, we reconstruct TWS anomalies (TWSA) for the period Jan 1960 to Dec 2022 thereby filling data gaps between GRACE and GRACE-FO missions as well as generating a complete dataset for the pre-GRACE era. The workflow involves identifying optimal predictors from land surface model (LSM) outputs, meteorological variables, and climatic indices using a novel Bayesian Network (BN) technique for grid-based TWSA simulations. Climate indices, like the Oceanic Niño Index and Dipole Mode Index, are selected as optimal predictors for a large number of grids globally, along with TWSA from LSM outputs. The most effective machine learning (ML) algorithms among Convolutional Neural Network (CNN), Support Vector Regression (SVR), Extra Trees Regressor (ETR), and Stacking Ensemble Regression (SER) models are evaluated at each grid location to achieve optimal reproducibility. Globally, ETR performs best for most of the grids which is also noticed at the river-basin scale, particularly for the Ganga-Brahmaputra-Meghana, Godavari, Krishna, Limpopo, and Nile river basins. The simulated TWSA (BNML_TWSA) outperformed the TWSA from LSM outputs when evaluated against GRACE datasets. Improvements are particularly noted in the river basins such as Godavari, Krishna, Danube, Amazon, etc., with median values of the correlation coefficient, Nash-Sutcliffe efficiency, and RMSE for all grids in Godavari, India, being 0.927, 0.839, and 63.7 mm respectively. A comparison with TWSA reconstructed in recent studies indicates that the proposed BNML_TWSA outperforms them globally as well as for all the 11 major river basins examined. The presented dataset is published at https://doi.org/10.6084/m9.figshare.25376695 (Mandal et al., 2024) and updates will be published when needed.
- 1 August 2023
- Published by Taylor & Francis in Hydrological Sciences Journal
- Vol. 68 (12) , 1693-1710
Contribution of climate and catchment characteristics to runoff variations in Indian river basins: a climate elasticity approach
- 1 August 2023
- Published by Taylor & Francis in Hydrological Sciences Journal
- Vol. 68 (12) , 1693-1710
Contribution of climate and catchment characteristics to runoff variations in Indian river basins: a climate elasticity approach
- Open Access
- 15 May 2023
- Published by Copernicus GmbH
Comparison of Neural network based and Kernel based Machine Learning approaches for daily forecasting of reference evapotranspiration in data scarce regions
Efficient estimation and forecast of reference evapotranspiration (ETO) is crucial for water resources management and for developing an efficient irrigation practice that will help better utilization of scanty water resources. This is a more challenging task in data scarce regions. This study aimed at multi-step ahead prediction of ETO across different cropping seasons and agro-climatic regions in India with six Machine learning (ML) based techniques using globally available fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) gridded reanalysis products (ERA5). For real-time, one-day, two-day, and seven-day ahead prediction of ETO, this study evaluates and compares the capability and prediction accuracy of deep learning algorithms, i.e., Support Vector Regression (SVR), Multivariate Adaptive Regression Splines (MARS), Random Forest (RF), Multi-Layer Perceptron (MLP), one-dimensional Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). ML based models were developed using meteorological observations along with ERA5 inputs at three meteorological stations: Nagpur, Hyderabad and Bhubaneswar. The results indicate that MLP, SVR and CNN outperform other ML algorithms. High performing models, one each (MLP and SVR) from neural network-based models and kernel-based models respectively, are further utilized to scale up the analysis for gridwise ETO forecasting across the whole of India. The Global Land Evaporation Amsterdam Model (GLEAM) dataset has been used as reference to evaluate gridwise ETO forecasts. ETO predicted by MLP model shows better agreement with GLEAM ETO values during the Rabi cropping season (October-March) (MAE = 0.103 mm/day and NRMSE = 3.9 %) than during the Kharif season (June-September) (MAE = 0.151 mm/day and NRMSE = 4.5 %). As expected, the accuracy of the models drops with increase in the prediction horizon from real-time to seven-day; for instance, with MLP, MAE = 0.146 mm/day, R2 = 0.955 for real-time and MAE = 0.173 mm/day, R2 = 0.939 for seven-day ahead prediction over arid agro-climatic zone during Rabi season. However, even the minimum forecast performance observed in the semi-arid tropics region during Rabi season is reasonably good (MAE = 0.396 mm/day, R2 = 0.704 for real-time evaluation and MAE = 0.445 mm/day, R2 = 0.56 for seven-day ahead). This strengthens the potential of the proposed models for multi-step ahead ETO forecasting across varied cropping seasons and agro-climatic regions without depending on meteorological station data.
- 15 May 2023
- Published by Copernicus GmbH
Open AccessComparison of Neural network based and Kernel based Machine Learning approaches for daily forecasting of reference evapotranspiration in data scarce regions
Efficient estimation and forecast of reference evapotranspiration (ETO) is crucial for water resources management and for developing an efficient irrigation practice that will help better utilization of scanty water resources. This is a more challenging task in data scarce regions. This study aimed at multi-step ahead prediction of ETO across different cropping seasons and agro-climatic regions in India with six Machine learning (ML) based techniques using globally available fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) gridded reanalysis products (ERA5). For real-time, one-day, two-day, and seven-day ahead prediction of ETO, this study evaluates and compares the capability and prediction accuracy of deep learning algorithms, i.e., Support Vector Regression (SVR), Multivariate Adaptive Regression Splines (MARS), Random Forest (RF), Multi-Layer Perceptron (MLP), one-dimensional Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). ML based models were developed using meteorological observations along with ERA5 inputs at three meteorological stations: Nagpur, Hyderabad and Bhubaneswar. The results indicate that MLP, SVR and CNN outperform other ML algorithms. High performing models, one each (MLP and SVR) from neural network-based models and kernel-based models respectively, are further utilized to scale up the analysis for gridwise ETO forecasting across the whole of India. The Global Land Evaporation Amsterdam Model (GLEAM) dataset has been used as reference to evaluate gridwise ETO forecasts. ETO predicted by MLP model shows better agreement with GLEAM ETO values during the Rabi cropping season (October-March) (MAE = 0.103 mm/day and NRMSE = 3.9 %) than during the Kharif season (June-September) (MAE = 0.151 mm/day and NRMSE = 4.5 %). As expected, the accuracy of the models drops with increase in the prediction horizon from real-time to seven-day; for instance, with MLP, MAE = 0.146 mm/day, R2 = 0.955 for real-time and MAE = 0.173 mm/day, R2 = 0.939 for seven-day ahead prediction over arid agro-climatic zone during Rabi season. However, even the minimum forecast performance observed in the semi-arid tropics region during Rabi season is reasonably good (MAE = 0.396 mm/day, R2 = 0.704 for real-time evaluation and MAE = 0.445 mm/day, R2 = 0.56 for seven-day ahead). This strengthens the potential of the proposed models for multi-step ahead ETO forecasting across varied cropping seasons and agro-climatic regions without depending on meteorological station data.