Analytical and Environmental Science Division & Centralized Instrument Facility, CSIR-Central Salt & Marine Chemicals Research Institute, Bhavnagar 364 002, Gujarat, India; Academy of Scientific and Innovative Research, Ghaziabad 201 002, Uttar Pradesh, India.
Analytical and Environmental Science Division & Centralized Instrument Facility, CSIR-Central Salt & Marine Chemicals Research Institute, Bhavnagar 364 002, Gujarat, India.
Mar Pollut Bull. 2023 May;190:114839. doi: 10.1016/j.marpolbul.2023.114839. Epub 2023 Mar 24.
Phytoplankton acts as carbon sinks due to photosynthetic efficacy and their diversity is expressed by SWDI (Shannon-Weaver Diversity Index), which depends on water quality parameters. The coastal water of Diu was studied for three seasons, and the relationship between different parameters and SWDI was established. Subsequently, an attempt was made to build up a prediction model of SWDI based on multilayer perceptron Artificial neural network (ANN) using the R programme. Analysis shows interrelationship between the water quality parameters and phytoplankton diversity is same in linear principal component analysis (PCA) and neural network model. Variations of different parameters depend on seasonal changes. The ANN model shows that ammonia and phosphate are key parameters that influence the SWDI of phytoplankton. Seasonal variation in SWDI is related to variation in water quality parameters, as explained by both ANN and PCA. Hence, the ANN model can be an important tool for coastal environmental interaction study.
浮游植物由于光合作用而成为碳汇，其多样性由 SWDI（香农-威弗多样性指数）表示，而 SWDI 又取决于水质参数。本文对迪奥的沿海水域进行了三个季节的研究，并建立了不同参数与 SWDI 之间的关系。随后，尝试使用 R 程序基于多层感知器人工神经网络 (ANN) 建立 SWDI 的预测模型。分析表明，水质参数与浮游植物多样性之间的相互关系在线性主成分分析 (PCA) 和神经网络模型中是相同的。不同参数的变化取决于季节变化。ANN 模型表明，氨和磷酸盐是影响浮游植物 SWDI 的关键参数。SWDI 的季节性变化与水质参数的变化有关，这一点可以通过 ANN 和 PCA 来解释。因此，ANN 模型可以成为海岸环境相互作用研究的重要工具。
