Projected Global Area Equipped for Irrigation Datasets during 2020-2100 under SSP scenarios
收藏资源简介:
1. Background 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. However, the predictive datasets of the area equipped for irrigation are still lacking. To address this gap, we provide the Projected Global Area Equipped for Irrigation Datasets (PGAEID), which provide spatially explicit estimates of Area Equipped for Irrigation (AEI) from 2020 to 2100 under three Shared Socioeconomic Pathway (SSP) scenarios: SSP1 (sustainable development), SSP2 (intermediate development), and SSP3 (regional rivalry), SSP4 (unequal development), and SSP5 (fossil-fueled development). 2. Methodology 2.1 Ensemble Machine Learning (EML) Framework Algorithms: Integrated six machine learning models: Multiple Linear Regression (MLR) Decision Trees (DT) Autoregressive Integrated Moving Average (ARIMA) Multi-Layer Perceptron (MLP) Radial Basis Function (RBF) Random Forests (RF) Training Data: Historical national irrigation records (FAO AQUASTAT, 1961–2015). Validation Metrics: Nash-Sutcliffe Efficiency (NSE): 0.98 Kling-Gupta Efficiency (KGE): 0.97 Mean Absolute Percentage Error (MAPE): 1.7% 2.2 Spatial Downscaling Baseline: FAO 2005 irrigation data combined with GMIA2005 gridded agricultural intensity maps. Dynamic Projection: Annual change rates applied to 5′ × 5′ grids under SSP-specific socioeconomic drivers. 3. Dataset Overview 3.1 Key Features Temporal Coverage: 2020–2100 (10-year intervals). Spatial Resolution: 5-arcminute (≈10 km at the equator). Scenarios: SSP1, SSP2, SSP3, SSP4, SSP5. Variables: area equipped for irrigation (103 ha/year). 3.2 Dataset Structure The dataset is provided as a compressed archive (PGAEID_Ver3.0.rar), containing: 1.Global_Area_Equipped_for_Irrigation_GeoTiff/ Subfolders: SSP1 SSP2 SSP3 SSP4 SSP5 File Format: GeoTIFF (45 files total). Naming Convention:AEI_[SSP]_[Year].tif Example: AEI_SSP1_2020.tif 2. National & Regional_AEI/ Shapefiles: National/regional area equipped for irrigation for 26 prediction units (2020–2100). Excel File: Global Area Equipped for Irrigation (2020-2100).xlsx. 3. Technical Annex.docx Detailed methodology, validation, and workflow documentation. 4. Applications This dataset supports: Earth System Simulation: Supporting irrigation parameterization in global climate and hydrological models. Water Resource Management: Assisting decision-makers in sustainable irrigation planning. Climate Change Adaptation: Providing insights into how irrigation practices evolve under different socioeconomic pathways. Environmental Conservation: Assessing the impact of irrigation on regional ecosystems. Note: Global aggregated totals of area equipped for irrigation derived from the 26 prediction units (country/regional scale) may exhibit minor discrepancies compared to sums calculated from the 5-arcminute gridded data (≈10 km resolution). Such differences stem from variations in spatial aggregation methods, file formats (vector vs. raster), and underlying data processing frameworks. Users may select the dataset best aligned with their analytical objectives: The country/region-based data (26 units) is recommended for national-scale analyses or policy evaluations requiring administrative boundaries. The 5-arcminute gridded data is preferable for spatially explicit modeling or subnational assessments. Both datasets maintain equivalent quality and methodological rigor; the choice depends on the desired spatial granularity and application context.
1. 研究背景 精准预测未来全球有效灌溉面积(Area Equipped for Irrigation, AEI),可为地球系统模拟、农业水资源管理、气候变化适应以及环境保护等领域提供核心支撑数据集,其重要性不言而喻。然而当前有效灌溉面积的预测数据集仍相对匮乏。为填补这一研究空白,本团队发布了全球未来有效灌溉面积预测数据集(Projected Global Area Equipped for Irrigation Datasets, PGAEID),该数据集针对5种共享社会经济路径(Shared Socioeconomic Pathway, SSP)情景下2020至2100年的有效灌溉面积开展空间显式估算,具体情景包括:SSP1(可持续发展路径)、SSP2(中等发展路径)、SSP3(区域竞争路径)、SSP4(不均衡发展路径)以及SSP5(化石燃料驱动发展路径)。 2. 研究方法 2.1 集成机器学习(Ensemble Machine Learning, EML)框架 算法设置:整合了6种机器学习模型,分别为:多元线性回归(Multiple Linear Regression, MLR)、决策树(Decision Trees, DT)、自回归积分滑动平均模型(Autoregressive Integrated Moving Average, ARIMA)、多层感知机(Multi-Layer Perceptron, MLP)、径向基函数(Radial Basis Function, RBF)以及随机森林(Random Forests, RF)。 训练数据:采用1961—2015年的历史全国灌溉记录(FAO AQUASTAT)。 验证指标: 纳什-舒特克利夫效率(Nash-Sutcliffe Efficiency, NSE):0.98; 克林-古普塔效率(Kling-Gupta Efficiency, KGE):0.97; 平均绝对百分比误差(Mean Absolute Percentage Error, MAPE):1.7%。 2.2 空间降尺度 基准数据:采用FAO 2005年灌溉数据结合GMIA2005格网化农业强度图谱。 动态投影:针对各SSP情景下的社会经济驱动因子,将年变化率应用于5′×5′的格网单元。 3. 数据集概览 3.1 核心特征 时间覆盖范围:2020—2100年,时间间隔为10年; 空间分辨率:5角分(赤道区域约合10公里); 情景类型:SSP1、SSP2、SSP3、SSP4、SSP5; 变量:有效灌溉面积(单位:10³公顷/年)。 3.2 数据集结构 本数据集以压缩归档文件(PGAEID_Ver3.0.rar)形式提供,包含以下内容: 1. Global_Area_Equipped_for_Irrigation_GeoTiff/ 子文件夹:SSP1、SSP2、SSP3、SSP4、SSP5; 文件格式:GeoTIFF(总计45个文件); 命名规则:AEI_[SSP]_[Year].tif; 示例:AEI_SSP1_2020.tif。 2. National & Regional_AEI/ 形状文件(Shapefile):针对26个预测单元的国家/区域尺度有效灌溉面积数据(2020—2100年); Excel文件:Global Area Equipped for Irrigation (2020-2100).xlsx。 3. Technical Annex.docx:包含详细的方法论说明、验证流程与工作流文档。 4. 应用场景 本数据集可支撑以下研究与应用方向: 地球系统模拟:为全球气候与水文模型中的灌溉参数化过程提供支撑; 水资源管理:辅助决策者制定可持续灌溉规划方案; 气候变化适应:揭示不同社会经济路径下灌溉实践的演变规律; 环境保护:评估灌溉活动对区域生态系统的影响。 注:由26个预测单元(国家/区域尺度)汇总得到的全球有效灌溉面积总估值,与基于5角分(约10公里分辨率)格网数据计算得到的总量可能存在小幅差异。此类差异源于空间聚合方法、文件格式(矢量与栅格的差异)以及底层数据处理框架的不同。用户可根据分析目标选择适配的数据集: - 若开展国家级分析或需要行政边界支撑的政策评估,推荐使用国家/区域尺度数据集(26个单元); - 若开展空间显式建模或次国家级评估,优先选用5角分格网数据。 两类数据集的质量与方法学严谨性保持一致,具体选择取决于所需的空间粒度与应用场景。



