A Review on Evapotranspiration Estimation in Agricultural Water Management: Past, Present, and Future

Spatiotemporal dynamics of ecosystem services in response to climate variability in Maze National Park and its environs, southwestern Ethiopia

Climate variability is one of the major factors affecting the supply of ecosystem services and the well-being of people who rely on them. Despite the substantial effects of climate variability on ecosystem goods and services, empirical researches on these effects are generally lacking. Thus, this study examines the spatiotemporal impacts of climate variability on selected ecosystem services in Maze National Park and its surroundings, in southwestern Ethiopia. We conducted climate trend and variability analysis by using the Mann-Kendall (MK) trend test, Sen's slope estimator, and innovative trend analysis (ITA). Relationships among ecosystem services and climate variables were evaluated using Pearson's correlation coefficient (r), while partial correlation was used to evaluate the relationship among key ecosystem services and potential evapotranspiration (PET). The MK tests show a decreasing trend for both mean annual and main rainy season rainfall, with Sen's slope (β) = -0.721 and β = -0.1.23, respectively. Whereas, the ITA method depicted a significant increase in the second rainy season rainfall (Slope(s) = 1.487), and the mean annual (s = 0.042), maximum (s = 0.024), and minimum (s = 0.060) temperature. Spatial correlations revealed significant positive relationships between ecosystem services and the mean annual rainfall and Normalized Difference Vegetation Index (NDVI), while negative correlations with the mean annual temperature. Additionally, temporal correlations highlighted positive relationships among key ecosystem services and the main rainy season rainfall. The maximum and minimum temperatures and ecosystem services were negatively correlated; whereas, there was strong negative correlations between annual (r = -0.929), main rainy season (r = -0.990), and second rainy season (r = -0.814) PET and food production. Thus, understanding the spatiotemporal variability of climate and the resulting impacts on ecosystem services helps decision-makers design ecosystem conservation and restoration strategies to increase the potential of the ecosystems to adapt to and mitigate the impacts of climate variability.

An evolutionary parsimonious approach to estimate daily reference evapotranspiration

The reference evapotranspiration (ETo) is an essential component in hydrological and ecological processes. The objective of this research is to develop an explicit model to estimate ETo only using commonly measurable meteorological parameters such as relative humidity, air temperature, and wind speed, where the measurements corresponding to solar radiation are omitted. The model was generated using Genetic Programming (GP), evaluated, and validated with reference data ETo using FAO56-PM. This reference data was obtained from different climates (warm-temperate and arid-warm) and latitudes, acquired from CIMIS stations in the state of California, United States, and the El Porvenir station in the state of Coahuila, located in north-central Mexico. After applying the proposed methodology, a total of 3754 results were generated, demonstrating a significant improvement in the estimation of ETo compared to the Hargreaves-Samani model. A particularly noteworthy result revealed that our approach outperformed the Hargreaves-Samani model in the training phase by 27%, and in the testing phase by 16%, on average. In order to achieve a generalized model, a dataset encompassing meteorological stations in two different climates (warm-temperate and arid-warm) and various latitudes was utilized. The obtained outcome unveiled a highly effective model for estimating ETo in diverse climatic contexts, eliminating the need for local adjustments. This model significantly surpassed the Hargreaves-Samani model, exhibiting superior performance by 17% during the training phase and 18% during the testing phase. These results conclusively underscore the capability of our approach to provide more accurate and reliable ETo estimates. These results conclusively underscore the capability of our approach to provide more accurate and reliable ETo estimates. Finally, to validate the model, four different datasets with climates similar to those used for model creation (warm-temperate, warm-arid) and different latitudes were employed. The validation stage results clearly indicate the superiority of our reference evapotranspiration ETo11 model over the Hargreaves-Samani model by 51% in warm-temperate climates. For the dataset with arid-warm climate, our model continued to show satisfactory results, surpassing the Hargreaves-Samani model by 8%. GP emerges as an innovative and effective alternative for simplified model development. This approach introduces a novel paradigm that facilitates the efficient development of models, standing out for its simplicity and effectiveness in generating solutions.

Estimating Groundwater Pumping for Irrigation: A Method Comparison

Effective groundwater management is critical to future environmental, ecological, and social sustainability and requires accurate estimates of groundwater withdrawals. Unfortunately, these estimates are not readily available in most areas due to physical, regulatory, and social challenges. Here, we compare four different approaches for estimating groundwater withdrawals for agricultural irrigation. We apply these methods in a groundwater-irrigated region in the state of Kansas, USA, where high-quality groundwater withdrawal data are available for evaluation. The four methods represent a broad spectrum of approaches: (1) the hydrologically-based Water Table Fluctuation method (WTFM); (2) the demand-based SALUS crop model; (3) estimates based on satellite-derived evapotranspiration (ET) data from OpenET; and (4) a landscape hydrology model which integrates hydrologic- and demand-based approaches. The applicability of each approach varies based on data availability, spatial and temporal resolution, and accuracy of predictions. In general, our results indicate that all approaches reasonably estimate groundwater withdrawals in our region, however, the type and amount of data required for accurate estimates and the computational requirements vary among approaches. For example, WTFM requires accurate groundwater levels, specific yield, and recharge data, whereas the SALUS crop model requires adequate information about crop type, land use, and weather. This variability highlights the difficulty in identifying what data, and how much, are necessary for a reasonable groundwater withdrawal estimate, and suggests that data availability should drive the choice of approach. Overall, our findings will help practitioners evaluate the strengths and weaknesses of different approaches and select the appropriate approach for their application.