<p>The present study aims to evaluate the performance of a hydrological model to si... more <p>The present study aims to evaluate the performance of a hydrological model to simulate spring runoff processes, analyze the effect of snowmelt on seasonal flow, and identify the snowmelt contribution rate based on the snow water equivalent (SWE) in the Moroccan High Atlas watersheds.</p> <p>The main objective of this study is to evaluate the daily SWE in a poorly instrumented mountainous watershed using an improved hydrological model. The model algorithm improvement is considered an essential approach for better understanding the initial basin conditions that influence these hydrogeological behaviors. For this purpose, a seasonal analysis was performed to select flood events that reproduce this phenomenon.</p> <p>To this end, the calibration has been done by forcing the model with rainfall, runoff, temperature, and snow water equivalent (SWE), with an amelioration of the model algorithm. Interestingly enough, this improvement achieved 13% based on the Nash-Sutcliffe efficiency coefficients. Hence, the spring event flows were influenced by the snowmelt process, these results will have direct implications for flood event replication modeling and flood forecasting in these regions.</p> <p>The study demonstrates that this region is sensitive to the seasonal effect of snowmelt. Therefore, it is essential to take into account the contribution of snow in hydrological studies developed at the level of the Moroccan High Atlas mountainous watersheds. This approach is a great challenge that will improve the reproduction of seasonal flood events and allow a better forecast of flood events to reduce the uncertainties and risks of flooding in mountainous basin areas facing the same climate conditions.</p> <p><strong>Keywords:</strong> Precipitation, SWE, Hydrological modeling, Calibration, Mediterranean climate, flood events, Zat basin.</p>
Assessing the right amount of water needs for a specific crop is a key task for farmers and agron... more Assessing the right amount of water needs for a specific crop is a key task for farmers and agronomists to achieve efficient and optimal irrigation scheduling, and then an optimal crop yield. To this end, the reference evapotranspiration (ET0) was developed. It represents the atmospheric evaporation demand, and therefore an important variable for irrigation management. In this regard, several methods such as the FAO’s Penman-Monteith and Hargreaves have been used to model and estimate ET0. These methods use climatic parameters data for calculation procedures such as solar net radiation (Rn), saturation vapour pressure(es), and min-max air temperatures or a combination of them. In this paper, we investigated two proposed data-driven methods to predict ET0 values in a semi-arid region in Morocco. The first approach is based on forecasting techniques and the second one uses end-to-end modeling of ET0 based on meteorological data and machine learning models. The feature selection and engineering results show that solar global radiation (Rg) and mean air temperature (Ta) have a significance of more than 87% as relevant predictors features for the ET0. We then used them as input to machine learning regression models. Regression evaluation metrics showed that The XGboost regressor model performs well in both cross-validation with R2=0.93 in the first fold, and in hold-out validation with R2=0.92 and RMSE=0.55. As a final step, we compared the univariate time series forecasting of ET0 using the Facebook Prophet model versus the machine learning modeling method that we proposed. As goodness-of-fit measures, forecasting using machine learning modeling of ET0 showed better results in terms of both R2 and RMSE.
Water scarcity is a major problem in the arid and semi-arid areas of Morocco, where irrigation is... more Water scarcity is a major problem in the arid and semi-arid areas of Morocco, where irrigation is essential for agriculture. Crop growth models can enhance water use efficiency, thus providing an economic benefit while reducing pressure on water resources. In this study, we evaluated the modeling performance of the DSSAT-CERES-Wheat model in estimating Evapotranspiration (ETa), Total soil water (TSW), Grain yield, Tops weight and phenological stages of winter wheat in the semi-arid region of Tensift Al Haouz, Marrakech. The simulation was performed at a daily time step during two successive growing seasons 2002/2003 and 2003/2004. The model calibration was done firstly on two fields and ETa, TSW phenological stages, and productive variables were calibrated after the comparison of the simulated and observed data. Afterward, the validation was performed on four fields during the two growing seasons. The results showed that the model simulates reasonably good Grain yield, Tops weight and phenological stages. Moreover, The average values of RMSE between observed and measured ETa, TSW, Grain yield and Tops weight were respectively, 0.70mm/day, 25mm, 0.6 t/ha and 2 t/ha for the validation fields. Statistical parameters like R2, d, and NRMSE were additionally used and showed that the model simulates acceptably the above-mentioned parameters. Furthermore, The Penman-Monteith FAO56 and Priestley and Taylor Evapotranspiration simulation methods were compared, the average values of d and R2 were respectively 0.85, 0.70 for the Penman-Monteith method, and 0.80, 0.65 for the Priestley and Taylor method. Thus, the DSSAT model can be considered a useful tool for monitoring the management of wheat in arid and semi-arid regions.Keywords: DSSAT, wheat, irrigation, water scarcity, crop model
In recent decades, climate change has led to a sharp increase in water demand. Particularly in ag... more In recent decades, climate change has led to a sharp increase in water demand. Particularly in agriculture, this has put a great strain on already scarce water resources, increased the need for irrigation water, and led to overuse of groundwater. Therefore, sustainable management of water resources while maintaining good agricultural yield by monitoring crop water status is necessary for sustainable and rational management of these resources, especially in arid and semi-arid regions. For this purpose, a detailed knowledge of the different processes describing the diurnal water cycle of plants in a large area is essential. However, micrometeorological or physiological experimental measurements and their partitioning are laborious to perform and not very representative of large areas.In this regard, remote sensing is a particularly suitable tool for monitoring agricultural areas because of its global and repeated observation. Several studies have highlighted the sensitivity of radar data to vegetation water content especially over the rainforest with spatial scatterometers that observe differences between morning and evening acquisitions. On the other hand, in situ radar experiments with high temporal frequency have made it possible to analyze radar responses over tropical and boreal forests.This study relates to a similar experiment conducted on an olive orchard located in the semi-arid Mediterranean region of Chichaoua in central Morocco. It allows the acquisition of in situ C-band radar measurements in crop fields, which are acquired continuously, from a tower-based radar system, with a time step of 15 minutes.The temporal evolution of the interferometric coherence r is analyzed on different baselines Dt, ranging from 15 minutes to 30 days, for the main physiological stages of the olive tree. Four different two-month periods, from December 2020 to November 2022, are chosen as the main physiological stages based on field observations.The obtained results of r, especially at 15-min min-steps, show a global behavior similar to that observed in tropical and boreal forests: high values (r ≈1) are observed during the night (weak wind, vegetation resting), then a decrease/increase during the day mainly anti-symetric to the wind cycle. As over boreal and tropical forest, a decrease in r is observed before the wind picks up, with is time coincident with sap flows and ETR variations, traducing its sensitivity to water plant content.Results show that over olive orchard, the r diurnal cycle is less marked than over boreal and tropical forests, due to lower ETR rates and certainly due to a significant soil contribution over this less dense vegetation layer. Furthermore, r values decrease when temporal baselines increase, but values are still meaningful for Dt = 6 days (r = 0.3 compared to 0.6 for Dt = 15 min. for the summer period), available with Sentinel-1 missions.The present study provides particularly interesting results confirming the sensitivity of C-band coherence to vegetation water status, especially in the early morning. Further work needs to be pursued to verify if we are able to detect the water stress of these plants in semi-arid areas such as Chichaoua through coherence.
<p>Accurate measurement of precipitation is very important ... more <p>Accurate measurement of precipitation is very important for flood forecasting, hydrological modeling, and estimation of the water balance of any basin. The lack of a weather monitoring network is an obstacle to the accurate measurement of precipitation.</p><p>In most of the Moroccan High Atlas Mountains regions, ground observation stations are still unreliable and difficult to access due to several parameters, such as a large spatial and temporal variation of rainfall and ruggedness of topography, which lead to irregularity and scarcity of measuring stations. This area is characterized by arid and semi-arid climates where generally occurred a few rainy days but have experienced significant flash floods.</p><p>Consequently, floods are causing extended damages to the population and infrastructures every year. However, research on hydrological processes is limited due to the irregularity of the gauge station network and the large number of gaps frequently observed in the rainfall and runoff data acquired from the gauge stations. Remote sensing precipitation data with high spatial and temporal resolution are a potential alternative to gauged precipitation data.</p><p>This study evaluates the performance of the two satellite products: the Tropical Rainfall Measuring Mission (TRMM 3B43V7) Multi-satellite Precipitation Analysis (TMPA) and the Integrated Multi-satellite Retrievals for GPM (IMERG V06) (SPPs) to observed rainfall, at different time scales (daily, monthly, and annual) from 1 September 2000 to 31 August 2017 over the Ghdat watershed, with different statistical indices and hydrological assessment, to evaluate the reliability of these (SPPs) data to reproduce rainfall events by implementing them in a hydrological model, to determine their ability to detect all types of rainfall events.</p><p>Daily, monthly, and annual rainfall measurements were validated using widely used statistical measures (CC, RMSE, MAE, Bias, Nash, POD, FAR, FBI and ETS).</p><p>The results showed that: (1) The correlation between satellite precipitation data and rainfall precipitation demonstrated a high correlation on all daily, monthly, and annual scales. (2) The product (TRMM 3B42V7) exhibits better quality in terms of correlation on the monthly and annual scale, while the (GPM IMERG V06) product shows a high correlation on the daily scale compared to the measurements of the gauges. (3) The (GPM IMERG V06) product has better performance regarding the precipitation detection capability, compared to the (TRMM 3B42V7) product which could detect only tiny precipitation events, but not able to capture moderate or strong precipitation events. (4) Flood events can be simulated with the hydrological model using both observed precipitation data and satellite data with the Nash – Sutcliffe model efficiency coefficient (NSE) ranging from 0.65 to 0.90.</p><p>According to the results of this study, we concluded that (TRMM 3B42V7) and (GPM IMERG V06) satellite precipitation products can be used for flood modeling and water resource management, particularly in the semi-arid and Mediterranean region.</p>
<p>Mediterranean mountainous regions are strongly affected by flash flood e... more <p>Mediterranean mountainous regions are strongly affected by flash flood events causing many damages. The vulnerability to flooding in the Moroccan High Atlas, especially in the Tensift basin, has been increasing over the last decades. Rainfall-runoff models can be very useful for flash flood forecasting. However, event-based models require a reduction of their uncertainties related to the estimation of initial moisture conditions before a flood event. Soil moisture may strongly modulate the magnitude of floods and is thus a critical parameter to be considered in flood modeling.</p><p>Indeed, several studies have assimilated satellite soil moisture observations into rainfall-runoff models to improve their flood forecasting capabilities.</p><p>In order to have a better representation of the watershed states which leads to a better estimation of the streamflow. By exploiting the strong physical connection between soil moisture dynamics and precipitation, it has been shown that satellite soil moisture observations can also be used to improve the quality of precipitation observations.</p><p>The aim of this study is to compare daily soil moisture measurements obtained by time domain reflectometry (TDR) at Sidi Rahal station with satellite soil moisture products (European Space Agency Climate Change Initiative, ESA-CCI), in order to estimate the initial soil moisture conditions for each event. The systematic bias between soil moisture products and in situ measurements was corrected using a bias correction method. The correlations between soil moisture products and in situ observations are about 0.77 after the correction.  </p><p>However, a modeling approach based on rainfall-runoff observations of 30 sample flood events have been applied, from (2011 to 2018), in the Ghdat basin were extracted and modeled by an event-based rainfall-runoff model (HEC-HMS) which is based on the Soil Conservation Service (SCS-CN), loss model, and a Clark unit hydrograph was developed for simulation and calibration of the 10-minute rainfall runoff.</p><p>A similar approach could be implemented in other watersheds in this region for further operational purposes. This method is very satisfactory for reproducing rainfall-runoff events in this small Mediterranean mountainous watershed, the same approach could be implemented in other watersheds in this region. The results of this study indicate that the remote sensing data are theoretically useful for estimating soil moisture conditions in data-sparse watersheds in arid Mediterranean regions.</p><p><strong><span>Keywords: </span></strong><span>Soil moisture; Floods; Remote sensing; Hydrological modelling, CN method, Mediterranean basin.</span></p>
<p>The present study aims to evaluate the performance of a hydrological model to si... more <p>The present study aims to evaluate the performance of a hydrological model to simulate spring runoff processes, analyze the effect of snowmelt on seasonal flow, and identify the snowmelt contribution rate based on the snow water equivalent (SWE) in the Moroccan High Atlas watersheds.</p> <p>The main objective of this study is to evaluate the daily SWE in a poorly instrumented mountainous watershed using an improved hydrological model. The model algorithm improvement is considered an essential approach for better understanding the initial basin conditions that influence these hydrogeological behaviors. For this purpose, a seasonal analysis was performed to select flood events that reproduce this phenomenon.</p> <p>To this end, the calibration has been done by forcing the model with rainfall, runoff, temperature, and snow water equivalent (SWE), with an amelioration of the model algorithm. Interestingly enough, this improvement achieved 13% based on the Nash-Sutcliffe efficiency coefficients. Hence, the spring event flows were influenced by the snowmelt process, these results will have direct implications for flood event replication modeling and flood forecasting in these regions.</p> <p>The study demonstrates that this region is sensitive to the seasonal effect of snowmelt. Therefore, it is essential to take into account the contribution of snow in hydrological studies developed at the level of the Moroccan High Atlas mountainous watersheds. This approach is a great challenge that will improve the reproduction of seasonal flood events and allow a better forecast of flood events to reduce the uncertainties and risks of flooding in mountainous basin areas facing the same climate conditions.</p> <p><strong>Keywords:</strong> Precipitation, SWE, Hydrological modeling, Calibration, Mediterranean climate, flood events, Zat basin.</p>
Assessing the right amount of water needs for a specific crop is a key task for farmers and agron... more Assessing the right amount of water needs for a specific crop is a key task for farmers and agronomists to achieve efficient and optimal irrigation scheduling, and then an optimal crop yield. To this end, the reference evapotranspiration (ET0) was developed. It represents the atmospheric evaporation demand, and therefore an important variable for irrigation management. In this regard, several methods such as the FAO’s Penman-Monteith and Hargreaves have been used to model and estimate ET0. These methods use climatic parameters data for calculation procedures such as solar net radiation (Rn), saturation vapour pressure(es), and min-max air temperatures or a combination of them. In this paper, we investigated two proposed data-driven methods to predict ET0 values in a semi-arid region in Morocco. The first approach is based on forecasting techniques and the second one uses end-to-end modeling of ET0 based on meteorological data and machine learning models. The feature selection and engineering results show that solar global radiation (Rg) and mean air temperature (Ta) have a significance of more than 87% as relevant predictors features for the ET0. We then used them as input to machine learning regression models. Regression evaluation metrics showed that The XGboost regressor model performs well in both cross-validation with R2=0.93 in the first fold, and in hold-out validation with R2=0.92 and RMSE=0.55. As a final step, we compared the univariate time series forecasting of ET0 using the Facebook Prophet model versus the machine learning modeling method that we proposed. As goodness-of-fit measures, forecasting using machine learning modeling of ET0 showed better results in terms of both R2 and RMSE.
Water scarcity is a major problem in the arid and semi-arid areas of Morocco, where irrigation is... more Water scarcity is a major problem in the arid and semi-arid areas of Morocco, where irrigation is essential for agriculture. Crop growth models can enhance water use efficiency, thus providing an economic benefit while reducing pressure on water resources. In this study, we evaluated the modeling performance of the DSSAT-CERES-Wheat model in estimating Evapotranspiration (ETa), Total soil water (TSW), Grain yield, Tops weight and phenological stages of winter wheat in the semi-arid region of Tensift Al Haouz, Marrakech. The simulation was performed at a daily time step during two successive growing seasons 2002/2003 and 2003/2004. The model calibration was done firstly on two fields and ETa, TSW phenological stages, and productive variables were calibrated after the comparison of the simulated and observed data. Afterward, the validation was performed on four fields during the two growing seasons. The results showed that the model simulates reasonably good Grain yield, Tops weight and phenological stages. Moreover, The average values of RMSE between observed and measured ETa, TSW, Grain yield and Tops weight were respectively, 0.70mm/day, 25mm, 0.6 t/ha and 2 t/ha for the validation fields. Statistical parameters like R2, d, and NRMSE were additionally used and showed that the model simulates acceptably the above-mentioned parameters. Furthermore, The Penman-Monteith FAO56 and Priestley and Taylor Evapotranspiration simulation methods were compared, the average values of d and R2 were respectively 0.85, 0.70 for the Penman-Monteith method, and 0.80, 0.65 for the Priestley and Taylor method. Thus, the DSSAT model can be considered a useful tool for monitoring the management of wheat in arid and semi-arid regions.Keywords: DSSAT, wheat, irrigation, water scarcity, crop model
In recent decades, climate change has led to a sharp increase in water demand. Particularly in ag... more In recent decades, climate change has led to a sharp increase in water demand. Particularly in agriculture, this has put a great strain on already scarce water resources, increased the need for irrigation water, and led to overuse of groundwater. Therefore, sustainable management of water resources while maintaining good agricultural yield by monitoring crop water status is necessary for sustainable and rational management of these resources, especially in arid and semi-arid regions. For this purpose, a detailed knowledge of the different processes describing the diurnal water cycle of plants in a large area is essential. However, micrometeorological or physiological experimental measurements and their partitioning are laborious to perform and not very representative of large areas.In this regard, remote sensing is a particularly suitable tool for monitoring agricultural areas because of its global and repeated observation. Several studies have highlighted the sensitivity of radar data to vegetation water content especially over the rainforest with spatial scatterometers that observe differences between morning and evening acquisitions. On the other hand, in situ radar experiments with high temporal frequency have made it possible to analyze radar responses over tropical and boreal forests.This study relates to a similar experiment conducted on an olive orchard located in the semi-arid Mediterranean region of Chichaoua in central Morocco. It allows the acquisition of in situ C-band radar measurements in crop fields, which are acquired continuously, from a tower-based radar system, with a time step of 15 minutes.The temporal evolution of the interferometric coherence r is analyzed on different baselines Dt, ranging from 15 minutes to 30 days, for the main physiological stages of the olive tree. Four different two-month periods, from December 2020 to November 2022, are chosen as the main physiological stages based on field observations.The obtained results of r, especially at 15-min min-steps, show a global behavior similar to that observed in tropical and boreal forests: high values (r ≈1) are observed during the night (weak wind, vegetation resting), then a decrease/increase during the day mainly anti-symetric to the wind cycle. As over boreal and tropical forest, a decrease in r is observed before the wind picks up, with is time coincident with sap flows and ETR variations, traducing its sensitivity to water plant content.Results show that over olive orchard, the r diurnal cycle is less marked than over boreal and tropical forests, due to lower ETR rates and certainly due to a significant soil contribution over this less dense vegetation layer. Furthermore, r values decrease when temporal baselines increase, but values are still meaningful for Dt = 6 days (r = 0.3 compared to 0.6 for Dt = 15 min. for the summer period), available with Sentinel-1 missions.The present study provides particularly interesting results confirming the sensitivity of C-band coherence to vegetation water status, especially in the early morning. Further work needs to be pursued to verify if we are able to detect the water stress of these plants in semi-arid areas such as Chichaoua through coherence.
<p>Accurate measurement of precipitation is very important ... more <p>Accurate measurement of precipitation is very important for flood forecasting, hydrological modeling, and estimation of the water balance of any basin. The lack of a weather monitoring network is an obstacle to the accurate measurement of precipitation.</p><p>In most of the Moroccan High Atlas Mountains regions, ground observation stations are still unreliable and difficult to access due to several parameters, such as a large spatial and temporal variation of rainfall and ruggedness of topography, which lead to irregularity and scarcity of measuring stations. This area is characterized by arid and semi-arid climates where generally occurred a few rainy days but have experienced significant flash floods.</p><p>Consequently, floods are causing extended damages to the population and infrastructures every year. However, research on hydrological processes is limited due to the irregularity of the gauge station network and the large number of gaps frequently observed in the rainfall and runoff data acquired from the gauge stations. Remote sensing precipitation data with high spatial and temporal resolution are a potential alternative to gauged precipitation data.</p><p>This study evaluates the performance of the two satellite products: the Tropical Rainfall Measuring Mission (TRMM 3B43V7) Multi-satellite Precipitation Analysis (TMPA) and the Integrated Multi-satellite Retrievals for GPM (IMERG V06) (SPPs) to observed rainfall, at different time scales (daily, monthly, and annual) from 1 September 2000 to 31 August 2017 over the Ghdat watershed, with different statistical indices and hydrological assessment, to evaluate the reliability of these (SPPs) data to reproduce rainfall events by implementing them in a hydrological model, to determine their ability to detect all types of rainfall events.</p><p>Daily, monthly, and annual rainfall measurements were validated using widely used statistical measures (CC, RMSE, MAE, Bias, Nash, POD, FAR, FBI and ETS).</p><p>The results showed that: (1) The correlation between satellite precipitation data and rainfall precipitation demonstrated a high correlation on all daily, monthly, and annual scales. (2) The product (TRMM 3B42V7) exhibits better quality in terms of correlation on the monthly and annual scale, while the (GPM IMERG V06) product shows a high correlation on the daily scale compared to the measurements of the gauges. (3) The (GPM IMERG V06) product has better performance regarding the precipitation detection capability, compared to the (TRMM 3B42V7) product which could detect only tiny precipitation events, but not able to capture moderate or strong precipitation events. (4) Flood events can be simulated with the hydrological model using both observed precipitation data and satellite data with the Nash – Sutcliffe model efficiency coefficient (NSE) ranging from 0.65 to 0.90.</p><p>According to the results of this study, we concluded that (TRMM 3B42V7) and (GPM IMERG V06) satellite precipitation products can be used for flood modeling and water resource management, particularly in the semi-arid and Mediterranean region.</p>
<p>Mediterranean mountainous regions are strongly affected by flash flood e... more <p>Mediterranean mountainous regions are strongly affected by flash flood events causing many damages. The vulnerability to flooding in the Moroccan High Atlas, especially in the Tensift basin, has been increasing over the last decades. Rainfall-runoff models can be very useful for flash flood forecasting. However, event-based models require a reduction of their uncertainties related to the estimation of initial moisture conditions before a flood event. Soil moisture may strongly modulate the magnitude of floods and is thus a critical parameter to be considered in flood modeling.</p><p>Indeed, several studies have assimilated satellite soil moisture observations into rainfall-runoff models to improve their flood forecasting capabilities.</p><p>In order to have a better representation of the watershed states which leads to a better estimation of the streamflow. By exploiting the strong physical connection between soil moisture dynamics and precipitation, it has been shown that satellite soil moisture observations can also be used to improve the quality of precipitation observations.</p><p>The aim of this study is to compare daily soil moisture measurements obtained by time domain reflectometry (TDR) at Sidi Rahal station with satellite soil moisture products (European Space Agency Climate Change Initiative, ESA-CCI), in order to estimate the initial soil moisture conditions for each event. The systematic bias between soil moisture products and in situ measurements was corrected using a bias correction method. The correlations between soil moisture products and in situ observations are about 0.77 after the correction.  </p><p>However, a modeling approach based on rainfall-runoff observations of 30 sample flood events have been applied, from (2011 to 2018), in the Ghdat basin were extracted and modeled by an event-based rainfall-runoff model (HEC-HMS) which is based on the Soil Conservation Service (SCS-CN), loss model, and a Clark unit hydrograph was developed for simulation and calibration of the 10-minute rainfall runoff.</p><p>A similar approach could be implemented in other watersheds in this region for further operational purposes. This method is very satisfactory for reproducing rainfall-runoff events in this small Mediterranean mountainous watershed, the same approach could be implemented in other watersheds in this region. The results of this study indicate that the remote sensing data are theoretically useful for estimating soil moisture conditions in data-sparse watersheds in arid Mediterranean regions.</p><p><strong><span>Keywords: </span></strong><span>Soil moisture; Floods; Remote sensing; Hydrological modelling, CN method, Mediterranean basin.</span></p>
Uploads
Papers by Saïd Khabba