This study provides a new methodology for simulating environmental water stress index (EWSI) that addresses environmental droughts' stochastic nature at regional and local scales. The current research used a case study of the Upper Ewaso Ngiro river basin in Kenya that possesses regional disparities attributed to climatic, biophysical, and anthropogenic variables. A stochastic modelling approach that ensembled 4D Euclidean feature space algorithm, least-squares adjustment, and iterations integrated the four environmental droughts indicators (meteorological, agricultural, socio-economic, and hydrological) into a single multivariate index called EWSI. The correlation between the simulated EWSI and initial reconnaissance drought index $(RD{I}_{\alpha }$) produced a correlation coefficient (r) of −0.93 and a p-value < 0.02. The correlation between EWSI and river discharge had a correlation coefficient of −0.89 and a p-value < 0.02. The assessment of severity revealed that 67–100% of the basin exhibited moderate to extreme environmental water stresses conditions between 1986 and 2018.

这项研究提供了一种模拟环境水分胁迫指数（EWSI）的新方法，该方法解决了区域和地方范围内环境干旱的随机性。当前的研究使用了肯尼亚上埃瓦索恩吉罗上游流域的案例研究，该流域具有因气候，生物物理和人为因素而引起的区域差异。一种将4D欧式特征空间算法，最小二乘平差和迭代相结合的随机建模方法，将四个环境干旱指标（气象，农业，社会经济和水文指标）整合到一个称为EWSI的单一多元指标中。模拟EWSI与初始勘测干旱指数的相关性。$（\mathrm{[R}d{\mathrm{一世}}_{\alpha }$）产生的相关系数（r）为-0.93，p值<0.02。EWSI与河流流量之间的相关系数为-0.89，p值<0.02。严重程度评估显示，在1986年至2018年期间，流域的67-100％表现出中度至极端的环境水分胁迫条件。