Projected Global Area Equipped for Irrigation Datasets during 2020-2100 under SSP scenarios

Accurately predicting the global area equipped for irrigation in the future is crucial for providing essential datasets relevant to fields such as earth system simulation, agricultural water resource management, climate change adaptation, and environmental conservation.

In this study, we developed a grid (5′×5′) dataset of global areas equipped for irrigation from 2020 to 2100 based on shared socioeconomic pathway (SSP) scenarios. This was achieved using an ensemble machine learning (EML) approach (Gao et al., 2024) that incorporates six algorithms: multiple linear regression (MLR), decision trees (DT), auto-regressive integrated moving average (ARIMA), multi-layer perceptron (MLP), radial basis function (RBF), and random forests (RFs).

Utilizing national irrigation records spanning from 1961 to 2015 (fao.org), our proposed EML approach was rigorously trained and validated against historical data with satisfactory accuracy.

The results indicate that the EML framework provides a robust and reliable method for irrigation prediction, consistently outperforming individual algorithms with high accuracy metrics—Nash-Sutcliffe efficiency of 0.98, Kling-Gupta efficiency of 0.97, and mean absolute percentage error of just 1.7%.

Under SSP1 and SSP2 scenarios, the global area equipped for irrigation is projected to decrease by approximately 7.13% and 9.35%, respectively, between 2020 and 2100; conversely, it may increase by about 5.26% under the SSP3 scenario. On average, India has the highest area equipped for irrigation globally, followed by China and the United States; Poland exhibits the lowest figures. Significant changes are anticipated in China's irrigated areas as well as those in India, Turkey, and Sub-Saharan Africa.

We contend that these Projected Global Area Equipped for Irrigation Datasets serve as a valuable complement to existing resources. They are expected not only to enhance our understanding of dynamics related to global irrigated areas across different socioeconomic pathways but also to provide support for pertinent research endeavors including earth system simulation, water resource management strategies, climate change adaptation efforts, and environmental conservation initiatives.

The Projected Global Area Equipped for Irrigation Datasets are packaged in a zip file named PGAEID.rar which contains three folders corresponding to each scenario—SSP1, SSP2—and one file titled 'Global Area Equipped for Irrigation (2020-2100).xlsx' upon extraction.

Among these files is 'Global Area Equipped for Irrigation (2020-2100).xlsx', which presents projections of global irrigated areas across various countries and regions under all three SSP scenarios from 2020 through 2100.

Each folder pertaining to SSP1, SSP2, or SSP3 includes gridded data representing global areas equipped for irrigation at ten-year intervals from 2020 until20300; each folder comprises nine GeoTIFF files resulting in a total of twenty-seven GeoTIFF files within this dataset.

Each GeoTIFF file follows a naming convention: AEI_[SSP scenario]_[year].tif—for instance, AEI_SSP1_2020.tif represents gridded projections specific to different countries/regions under the SSP1 scenario during the year 2020.

Other *.tif files can be identified similarly.