Insights From Intensive Stream Monitoring in an Eastern Mediterranean Agricultural Catchment Illuminate Anthropogenic Impact on Water Quality

Abstract

Degraded water quality is one of the well‐documented adverse effect of intensive agriculture on the riverine environment. It can result from nutrient leaching from agricultural fields, stream flow harvesting, irrigation with treated wastewater, full or partial damming, and cultivating the riparian zone. The weight of each cause on the regime water quality is only sometimes apparent. Thus, stream restoration planning requires understanding of the dynamics of water quality, and stakeholders' efforts do not address the right stressors on time. This is further exacerbated in Mediterranean agriculture catchments, which are prone to water scarcity, seasonal rainfall and streamflow variations. We therefore aim to evaluate how monitoring water quality parameters can help direct river restoration projects in Eastern Mediterranean watersheds. The present project focused on water quality dynamics along the main channel of the Nahalal stream and employed high spatiotemporal monitoring of basic water quality parameters: pH, electrical conductivity (EC), and macro‐nutrients in the context of (1) spatial gradients: Channel upstream vs. downstream; (2) seasonality: Base flow vs. flood water flows; and (3) hydraulic constraint: open vs. closed channel sluice gate. The results from the systematic stream grab water sampling, 34 times over two years at 13 locations, revealed increased concentrations at the downstream direction by dry‐wet conditions prior to the date of sampling. Contrary to expectations, higher concentrations of nitrates were found upstream near a spring—a source for base flow, than at the downstream reach, and the opposite trend was found for P concentrations. The results suggest that combined effects of several stressors control stream water quality dynamics: Riparian vegetation dominated by Phragmites australis, soil‐erosion processes, and water resources management. Lack of high spatiotemporal monitoring could result in misinterpretation of the intervention actions needed at the channel and watershed scales and may lead to poor initiation of self‐stream restoration ecology.