Predicting the effects of Controlled Tile Drainage on Nitrogen and Sediment Loads at the Watershed Scale using AnnAGNPS (Case study: Qharesu Watershed)

The entry of pollutants from agricultural land into water bodies is one of the most environmental issues connected to the tile drainage systems. Also leaching of drain water may cause the required nutrients for crop growth becomes out of reach. Thus, applying Best Management practies (BMPs) to relief the side effects of agricultural tile drainage systems is crucial. Over the past few decades, several BMPs have been introduced and studied with the aim of manage and control the volume and pollutants amount of tile drain water in agricultural lands, of which Controlled Tile Drainage (CTD) is one of them. Due to difficulty and costly nature of long-term monitoring programs, which is often used to assess impacts of applying BMPs on non-point source pollution, the AnnAGNPS model developed by USDA is a reliable alternative to simulate an effect of applying CTD at the watershed scale. AnnAGNPS, a comprehensive computational model that evolved from AGNPS, was designed for evaluating non-point pollutant loadings in agricultural watersheds and river basins. Briefly, AnnAGNPS is a distributed parameter simulation model that subdivides a watershed into small, homogeneous subwatersheds called “cells.” Cells are interconnected by stream channels called “reaches” In this study, an impact assessment of applying CTD on Total Suspended Sediment (TSS), Total Nitrogen (N total), Dissolved Nitrogen (N dissolved) and Surface Runoff were evaluated by AnnAGNPS at the Qharesu Watershed.

Qharesu watershed with an area of about 1670 square kilometers is one of the watersheds of Golestan province, which is surrounded by the Gorganrud watershed area from the north and east, Nekarud watershed in the south and the Gorgan Bay from the west. Finally, it is discharged into this bay. An average meteorological data of 5 statistical years of 2016 to 2020, used as a model input data in order to further adapt to the climatic conditions of the region in future, when tile drainage system is operated. The surface runoff data and suspended sediment load were also obtained from the Siah-Ab hydrometric station (outlet point of Qharesu watershed). Calibration and validation of the model performed by runoff and TSS observation data, in which the degree of confirmity between observation data and model output data performed by using three indices of Nash-Sutcliffe, RSR, and PBIAS.Two tile drain depth scenarios were examined in detail to mimic tile drainage control: The depth of 0.2m (CTD0.2) as an extensive controlled drainage and the depth of 0.6m (CTD0.6) as the recommended depth in the growing season for controlled drainage. The depth of the drainage base considered 1.1m as a reference condition as well.

The results showed that the model is reliable for simulating the studied parameters. The indices of E, RSR and PBIAS for runoff were 0.85, 0.48 and -2.6 for the calibration period and 0.76, 0.44 and -10.9 for the validation period respectively. These numbers were 0.73, 0.43 and -31.7 (calibration) and 0.66, 0.54 and -29.1 (validation) for sediment data.The AnnAGNPS model predicted a slight decrease in runoff and total nitrogen and a moderate decrease in dissolved nitrogen at the outlet of the Qharaesu River under the CTD0.6 scenario (during the growing season). Also, the increase in sediment load due to CTD0.6 was almost negligible. Based on the model results, Controlled drainage to a depth of approximately 0.6m below ground level is a suitable equilibrium point for reducing nitrogen and sediment loads at the watershed scale. Considering the importance of the outlet point of Gharesu, which is discharged to Gorgan bay, and since a significant part of tile drainage system in this watershed has not been impelemented yet, applying controlled tile drainage system is possible at a lower cost and more comfortable. Therefore, the possibility of applying CTD0.6 scenario should be on the agenda of authorities.