No abstract

Despite efforts to protect the hydrosystems from increasing pollution, nitrate (NO3−) remains a major groundwater pollutant worldwide, and determining its origin is still crucial and challenging. To disentangle the origins and fate of high NO3− (>900 mg/L) in the Sidi Bouzid North basin (Tunisia), a numerical groundwater flow model (MODFLOW-2005) and an advective particle tracking (MODPATH) have been combined with geostatistical analyses on groundwater quality and hydrogeological characterization. Correlations between chemical elements and Principal Component Analysis (PCA) suggested that groundwater quality was primarily controlled by evaporite dissolution and subsequently driven by processes like dedolomitization and ion exchange. PCA indicated that NO3− origin is linked to anthropic (unconfined aquifer) and geogenic (semi-confined aquifer) sources. To suggest the geogenic origin of NO3− in the semi-confined aquifer, the multi-aquifer groundwater flow system and the forward and backward particle tracking was simulated. The observed and calculated hydraulic heads displayed a good correlation (R2 of 0.93). The residence time of groundwater with high NO3− concentrations was more significant than the timespan during which chemical fertilizers were used, and urban settlements expansion began. This confirmed the natural origin of NO3− associated with pre-Triassic embankment landscapes and located on domed geomorphic surfaces with a gypsum, phosphate, or clay cover.

One of the major challenges in assessing groundwater vulnerability to pollution is the inadequate factors number and weight. Therefore, to carefully improve the assessment of groundwater vulnerability, a model independent of weight assignment errors was used. Moreover, the specific vulnerability index (SVI) of Sidi Bouzid North's groundwater was assessed in this study. Intrinsic vulnerability index (IVI) assessment was determined in the first step by the arithmetic mean calculation by the index overlay method (IOM) based on the D: Depth of aquifer; P: annual average Precipitation; L: Lithology of the vadose zone and S: percent of Slope (DPLS) model. Then SVI was assessed by linking new factors (LU and NO 3 − ) to IVI. Consequently, 83 samples were analysed for NO 3 − , showing high values that exceed 50 mg l¹. The spatial distribution of IVI shows three vulnerability classes in the study area, namely low (8%), moderate (15%)and high (77%). The evaluation of SVI based on the risks associated with the NO 3 − and LU factors using GIS indicates that about 95% of the total study area is classified with high SVI levels. It displayed a good correlation with NO 3 − and provided a high discretization of the groundwater risk from natural and anthropogenic pollution. This alarming situation suggests the necessity to apply water-saving irrigation action for adequate water management.

Floods frequently threaten villages near the Khazir River’s floodplains, causing crop losses and threatening residential areas. We used flood-related hydrological software, including WMS and HEC-HMS, to study this issue and determine how to reduce the recurrence of flooding. The software can be used to calculate a hydrograph of torrential flows in a river drainage basin and estimate the volume of torrential water and its flow rates on the Earth’s surface. The depth of rain has been evaluated and calculated in the SCS Unit Hydrograph for different return periods of 2, 5, 10, 20, 50, and 100 years. According to our study’s findings, the volume of the river’s drainage basin floods ranged between 29,680 and 2,229,200 m3, and the maximum flow value ranged between 10.4 and 66.4 m3/sec during various reference periods. To analyze and model the flood risks of the Khazir River, the HEC-RAS model was combined with the HEC-GeoRAS extension in ArcGIS. The floods were the focus of two study periods, 2013 and 2018, and were based on the digital elevation model and river discharge during the floods. According to the classification map of the flood depths, the areas of flood risk varied from low to very low (80.31%), medium (16.03%), and high to very high (3.8%). The analysis of the results revealed that the villages closest to the river’s mouth were more affected by the floods than other villages further downstream. HEC-HMS and HEC-RAS have been shown to have a strong correlation in evaluating flood risks and reliably forecasting future floods in the study area.