Water resources become more and more threatened by the increasing request related to the population growth, especially in agricultural regions. This article focused on the design of a geo-model monitoring network controlling groundwater quality in the fast-growing agricultural area of Souassi, Central eastern Tunisia. The studied area is known for its dry climate conditions and information scarcity while they are most needed. In this case, the integrated approach based on the Exploratory Factor Analysis coupled to the Confirmatory Factors Analysis and geostatistics were performed in order to detect the most relevant factors intervening in the groundwater quality and their spatial variability in the shallow aquifer of Souassi. This investigation is based on data set of nine parameters (EC, Ca, Mg, Na, K, HCO 3 , SO 4 , Cl and NO 3 ) determined in 30 samples from wells tapping the shallow aquifer of Souassi. The PCA results allowed illustrating that the groundwater quality in Souassi shallow aquifer is affected by both natural and anthropogenic factors. Developing the SEM led to identifying the most meaningful factors controlling the water mineralization. The geospatial interpolation using kriging technique enabled us to locate areas with high salinity values corresponding to the natural discharge zone of the aquifer as well as areas with high rate of nitrate coinciding essentially with olive fields in the central zone. Under limited database, the results allow highlighting the risk zones in the aquifer. This could assist in establishing sustainable agricultural activities while helping to solve environmental degradation of water resources in the study area.

中文翻译：

---

数据稀缺挑战下农业地区地下水资源可持续性的 EFA-CFA 综合方法：突尼斯中东部 Souassi 含水层案例研究

与人口增长相关的需求日益增加，特别是在农业地区，水资源越来越受到威胁。本文重点介绍了在突尼斯中东部 Souassi 快速增长的农业区控制地下水质量的地理模型监测网络的设计。研究区域以其干燥的气候条件和最需要的信息稀缺而闻名。在这种情况下，基于探索性因素分析与验证性因素分析和地质统计学相结合的综合方法被执行，以检测干预地下水质量的最相关因素及其在 Souassi 浅层含水层中的空间变异性。该调查基于九个参数（EC、Ca、Mg、Na、K、HCO 3 、SO 4 、Cl 和 NO 3 ) 在来自 Souassi 浅层含水层的井的 30 个样品中测定。PCA 结果表明，苏阿西浅层含水层的地下水质量受自然和人为因素的影响。开发 SEM 导致确定控制水矿化的最有意义的因素。使用克里金技术的地理空间插值使我们能够定位与含水层自然排放区相对应的高盐度值区域，以及与中央区域的橄榄田基本一致的高硝酸盐率区域。在有限的数据库下，结果可以突出含水层中的风险区域。这有助于建立可持续的农业活动，同时帮助解决研究区水资源的环境退化问题。