Abstract Holding proper ratio between groundwater extraction and surface water allocation in conjunctive use of water resources may be a desirable way for simultaneously controlling the aquifer sustainability and meeting the increasing agricultural water demands. This paper focuses on maximizing the agricultural net benefits (NB) while avoiding the vital groundwater resources to excessively withdraw when utilized to supply the water demands and provide desirable net benefits. For this purpose, the gravitational search algorithm (GSA) is employed as the optimizer and linked to the artificial neural network (ANN) as the simulator of the groundwater level (GWL) variations. The study area is the west of the irrigation network of Qazvin plain in Iran. This area is divided into two zones: low-GWL-drawdown zone (zone 1); and high-GWL-drawdown zone (zone 2). For each zone, nine scenarios are defined considering three climatic years each of which is assigned three different crop patterns, totally resulting in 18 scenarios. The results show that the simulation-optimization models can decrease the groundwater drawdown in all scenarios by 0.77–1.84 m in zone 1, and by 1.28–1.97 m in zone 2. Furthermore, by replacing the existing crop pattern with two other ones, NB is increased by 35.5–53.7% in the zone 1 and by 24.9–59.7% in the zone 2. Additionally, the net benefit per unit water consumption volume (NBPD) is increased by 18.9–32.2% in zone 1 and by 9–52.6% in zone 2.

中文翻译：

---

利用引力搜索算法评估经济-环境联合用水的优化方案

摘要 在水资源联合利用中保持地下水开采和地表水分配的适当比例可能是同时控制含水层可持续性和满足日益增长的农业用水需求的理想方式。本文的重点是最大限度地提高农业净收益 (NB)，同时避免重要的地下水资源在用于满足用水需求并提供理想的净收益时过度抽取。为此，采用重力搜索算法 (GSA) 作为优化器，并将人工神经网络 (ANN) 作为地下水位 (GWL) 变化的模拟器。研究区位于伊朗加兹温平原灌溉网络以西。该区域分为两个区域：低 GWL 水位下降区域（区域 1）；和高 GWL 下降区（区 2）。对于每个区域，考虑三个气候年定义了九个情景，每个气候年分配了三种不同的作物模式，总共产生 18 个情景。结果表明，模拟优化模型可以将所有情景下的地下水降幅在 1 区减少 0.77-1.84 m，在 2 区减少 1.28-1.97 m。此外，通过用另外两种模式替换现有的作物模式，NB 1 区增加 35.5-53.7%，2 区增加 24.9-59.7%。此外，单位用水量净收益 (NBPD) 在 1 区增加 18.9-32.2%，增加 9-52.6%区域 2 中的百分比。