Search

Olives constitute a frequently grown crop in semi-arid areas. Therefore, accurate quantification of evapotranspiration (ET) within olive groves is crucial to enhance agricultural water productivity and promote their resilience to water scarcity and future climate scenarios. In the present work, we assessed the accuracy of 3 versions of the Two-Source-Energy-Balance (TSEB) model, the first one “TSEB-SPT” using a standard Priestley-Taylor coefficient (αPT) to estimate the transpiration, the second one called “TSEB-CPT” constrained by a computed αPT using measured ET along with the equilibrium term, and the third one “TSEB-SM” where soil moisture is used as an additional constraint to improve the soil evaporation. The 3 models were applied over an irrigated olive orchard in the Tensift basin (Morocco) during two growing periods of 2003 and 2004. The comparison with ground-based flux measurements from Eddy-Covariance tower and sap flow data revealed that the TSEB-SPT model overestimates ET with an average relative error of 87% and a percentage bias of ‐78% during the two growing seasons. Conversely, TSEB-SM and TSEB-CPT improved ET estimates as compared to TSEB-SPT, with mean relative errors of 31% and 24% and an average percentage bias of 0.6% and ‐7.4%, respectively. For ET partitioning, TSEB-SM appears to be less effective in estimating transpiration, while the simulated transpiration by TSEB-CPT fits well the actual one with a root mean square error of 0.27 mm, mainly during the summer of 2003. These results open a path for future improvements: by reviewing the calibration procedure of αPT, and implementing alternative formulas to compute the evaporation, the TSEB-SM could be potentially a robust tool for monitoring the seasonal variation of ET and its partitioning over a heterogeneous canopy cover.

Africa is facing an urgent need to increase food production to meet increasing demands. Targeted investments in integrated agriculture and, water management systems are required to meet this challenge. However, there is a lack of comprehensive information on the potential applications of climate-smart agriculture (CSA). This paper reviews current crop modeling technologies and their applications within the scope of climate change and the CSA framework in Africa. It evaluates current research trends in various crop simulation models and suggest advanced approaches to improve crop and environmental assessment, crop management, and decision-making. A total of 140 relevant papers were considered. Results showed that 84% of studies used process-based models, with Maize being the most studied crop. Additionally, DSSAT crop models and analysis of variance models have the highest contribution of physical and empirical crop modeling studies respectively. Over 72% of studies have contributed to adaptation strategies and reducing yield gaps, while only 8% of studies have been conducted on climate change mitigation and their trade-offs with adaptation using crop models under CSA. To ensure food security through sustainable agricultural practices in Africa, there is crucial to implement CSA models with a focus on the climate change mitigation component.

The aim of this work is to use the AquaCrop model for irrigation and water productivity assessment of Table grapes in arid region of Mexico during 2005 and 2006 cropping seasons. The irrigation efficiency was investigated by comparing the irrigation scheduling design used by the farmer to the AquaCrop model recommendations. Data analysis showed that the farmer irrigates almost every day, which results in the water content in the root zone always exceeding the soil moisture at field capacity (FC). This generates substantial losses of water through deep percolation. By using the AquaCrop model, the optimization of irrigation water scheduling in order to avoid both water stress and deep percolation was about 547 mm and 510 mm, which it is about half of what was applied by the farmer (1006 mm and 930 mm) during 2005 and 2006, respectively. This large difference, lost through deep percolation, reduces the water productivity (WP) by about 45%.

The Plio-Quaternary aquifer in the EL-Haouz-Mejjate region of Morocco is critical for water supply, necessitating accurate characterization for sustainable management. This study pioneers machine learning (ML) and deep learning (DL) techniques to elucidate the aquifer’s properties. Supervised algorithms, including random forest, regression, support vector machines, Gaussian process regression and neural networks, are trained on available hydrogeological data. Diverse features capture complex input-output relationships to predict key hydrodynamic factors like hydraulic conductivity and transmissivity fields. Aquifer architecture attributes, including substratum depth, thickness, and height, are also estimated. Model outputs are validated with field measurements, demonstrating promising accuracy. Enhanced hydrodynamic insights improve the conceptual model and groundwater flow modeling confidence. Uncertainties are reduced through this data-driven approach, enabling optimized aquifer management. Overall, this work shows how useful it is to combine ML and DL with traditional hydrogeology in order to get a better understanding of complicated aquifer systems. The techniques pioneered provide a pathway for sustainable management of this vital water resource.

Accurately estimating latent heat flux (LE) is crucial for achieving efficiency in irrigation. It is a fundamental component in determining the actual evapotranspiration (ETa), which in turn, quantifies the amount of water lost that needs to be adequately compensated through irrigation. Empirical and physics-based models have extensive input data and site-specific limitations when estimating the LE. In contrast, the emergence of data-driven techniques combined with remote sensing has shown promising results for LE estimation with minimal and easy-to-obtain input data. This paper evaluates two machine learning-based approaches for estimating the LE. The first uses climate data, the Normalized Difference Vegetation Index (NDVI), and Land Surface Temperature (LST), while the second uses climate data combined with raw satellite bands. In-situ data were sourced from a flux station installed in our study area. The data include air temperatures (Ta), global solar radiation (Rg), and measured LE for the period 2015-2018. The study uses Landsat 8 as a remote sensing data source. At first, 12 raw available bands were downloaded. The LST is then derived from thermal bands using the Split Window algorithm (SW) and the NDVI from optical bands. During machine learning modeling, the CatBoost model is fed, trained, and evaluated using the two data combination approaches. Cross-validation of 3-folds gave an average RMSE of 27.54 W.nr2 using the first approach and 27.05 W.nr2 using the second approach. Results raise the question: Do we need additional computational layers when working with remote sensing products combined with machine learning? Future work is to generalize the approach and test it for other applications such as soil moisture retrieval, and yield prediction.

In this study, prominent dust source areas are identified along with their plume extent using high temporal frequency satellite observations. Hourly dust plume observations of the Dust Belt from geostationary‐orbit satellites are analyzed for the 2017‐12–2022‐11 period. To identify dust source areas and their extents, we back‐track plumes to their source, assessing source areas in terms of emission frequency, contribution, and plume extent patterns. This method advances over traditional source allocation techniques that rely on polar‐orbiting satellites based on a few daily passes and meteorological wind fields for backtracking. Our findings indicate that Boreal summer is the most intense season for most sources, except in the Southern Sahara, which experiences winterly winds. Our analysis also reveals significant contributions from regions within the Sahara that experience expansive but infrequent dust storms, highlighting the importance of considering both frequency and magnitude in understanding dust emissions. Plain Language Summary: This study focuses on the role of mineral dust, a significant climate constituent originating mainly from arid regions. Identifying where dust comes from and where it goes is essential for understanding how it affects the climate. To do so, we employ high temporal frequency satellite data to backtrack dust plumes to their sources, revealing key dust‐emitting regions and their seasonal variations. We also quantify each source area's frequency and magnitude of emissions and the spatial distribution of emanating dust plumes. Our findings show that summer is the peak season for most areas, except for the strongest dust sources, located in Southern Sahara, which peak in winter driven by seasonal winds. Additionally, some Saharan source areas exhibit severe storms but have been under‐emphasized because they were measured by the frequency of their storms. Key Points: High‐temporal‐resolution images from geostationary‐orbit satellites enable precise back‐tracking of dust source areasDust source areas are analyzed in terms of their emission frequency, contribution to the regional atmosphere, and corresponding plume extentSummer is the peak dust emission season for most sources in West Asia and Sahara and ephemeral lake beds are active during the dry season [ABSTRACT FROM AUTHOR]

Total nitrogen (TN) is a critical nutrient for plant growth, and its monitoring in agricultural soil is vital for farm managers. Traditional methods of estimating soil TN levels involve laborious and costly chemical analyses, especially when applied to large areas with multiple sampling points. Remote sensing offers a promising alternative for identifying, tracking, and mapping soil TN levels at various scales, including the field, landscape, and regional levels. Spaceborne hyperspectral sensing has shown effectiveness in reflecting soil TN levels. This study evaluates the efficiency of spectral reflectance at visible near-infrared (VNIR) and shortwave near-infrared (SWIR) regions to identify the most informative hyperspectral bands responding to the TN content in agricultural soil. In this context, we used PRISMA (PRecursore IperSpettrale della Missione Applicativa) hyperspectral imagery with ensemble learning modeling to identify N-specific absorption features. This ensemble consisted of three multivariate regression techniques, partial least square (PLSR), support vector regression (SVR), and Gaussian process regression (GPR) learners. The soil TN data (n = 803) were analyzed against a hyperspectral PRISMA imagery to perform spectral band selection. The 803 sampled data points were derived from open-access soil property and nutrient maps for Africa at a 30 m resolution over a bare agricultural field in southern Morocco. The ensemble learning strategy identified several bands in the SWIR in the regions of 900–1300 nm and 1900–2200 nm. The models achieved coefficient-of-determination values ranging from 0.63 to 0.73 and root-mean-square error values of 0.14 g/kg for PLSR, 0.11 g/kg for SVR, and 0.12 g/kg for GPR, which had been boosted to an R2 of 0.84, an RMSE of 0.08 g/kg, and an RPD of 2.53 by the ensemble, demonstrating the model's accuracy in predicting the soil TN content. These results underscore the potential for using spaceborne hyperspectral imagery for soil TN estimation, enabling the development of decision-support tools for variable-rate fertilization and advancing our understanding of soil spectral responses for improved soil management. [ABSTRACT FROM AUTHOR]