Optimization of ammonia nitrogen benchmarks and ecological risk assessment in monsoon freezing lakes based on species sensitivity distribution with Lake Chagan in northeastern China as an example

• Published in: Ecological indicators 2024-09, Vol.166, p.112346, Article 112346
• Main Authors: Lou, Yuqi, Bian, Jianming, Sun, Xiaoqing, Wang, Fan, Xu, Liwen, Sun, Guojing
• Format: Article
• Language: English
• Subjects: Ammonia nitrogen pollution; Ecological risk assessment; Species sensitivity distribution; Water-quality benchmarks
• ISSN: 1470-160X
• DOI: 10.1016/j.ecolind.2024.112346

•First application of SSD for ecological risk assessment in study area.•Using empirical values as water-quality benchmarks leads to overprotection.•GA reduces the uncertainty associated with the choice of model distribution. There is an urgent need to determine reasonable water-quality benchmarks and risk assessment results to accurately characterize the relationship between crop yield enhancement and ammonia nitrogen pollution caused by excessive nitrogen fertilizer application. Through the collection of monitoring data from 2016 to 2023 in Chagan Lake, the first development of ecological criteria for ammonia nitrogen and risk evaluation for Northeast China was carried out based on the secondary ecological risk models (the quotient value method and joint probability curve). In addition, the species sensitivity distribution (SSD) model is improved through the use of genetic algorithms to identify more suitable ammonia water-quality benchmarks for ecological risk assessment. The results showed that high ammonia nitrogen concentration in wet season is primarily influenced by agricultural irrigation and drainage activities, whereas the dry season is driven by low temperatures and the nitrogen and phosphorus concentrations. The biological benchmark of improved SSD model with a genetic algorithm has a maximum error reduction of 23.84 % compared with other distributions, which is more than twice that of the median water-quality benchmark under the traditional Class III water-quality standard and empirical formula. Risk evaluation using traditional water-quality standards leads to the overprotection of the biological environment. Evaluations using quotient value methods and joint probability curves yield consistent results. The severity of the ecological risk of ammonia nitrogen follows the order: wet season > flat season > dry season. Using a genetic algorithm to optimize SSD model can effectively reduce the uncertainty of the model, providing a more accurate and reasonable ammonia nitrogen water-quality benchmark. This study improves the evaluation of the ecological risk of ammonia nitrogen in Lake Chagan, provides ideas for reducing the uncertainty of the selection of ammonia nitrogen water-quality benchmarks, and effectively prevents the overprotection of the lake from ammonia nitrogen.