Investigation of the efficiency of support vector machine in predicting changes in water quality parameters (Case study: Choghakhor International Wetland)

,; ,; ,; ,; ,

Volume 10, Issue 1 (Summer 2020) J. Aqua. Eco 2020, 10(1): 23-34 | Back to browse issues page

‎ 20.1001.1.23222751.1399.10.1.3.9

Mendeley

Zotero

RefWorks

Investigation of the efficiency of support vector machine in predicting changes in water quality parameters (Case study: Choghakhor International Wetland). J. Aqua. Eco 2020; 10 (1) :23-34
URL: http://jae.hormozgan.ac.ir/article-1-862-en.html

Investigation of the efficiency of support vector machine in predicting changes in water quality parameters (Case study: Choghakhor International Wetland)

Abstract: (2927 Views)

Inland waters, such as wetlands, are considered to be sensitive ecosystems, and sustainable productivity can only be achieved by adopting an appropriate environmental approach. Estimating water quality parameters using models will reduce costs and provide faster access to water resources management. In this study, a 32-year support vector machine (SVM) model was used from 1985 to 2017 to simulate the chlorophyll a and water clarity (Secchi-disk depth) in Choghakhor international wetland. 70% of the data was used to train and the rest of the data was used for testing of the model. The determination coefficient (R²) and root mean square error (RMSE) in the SVM model were 0.93, 0.91, and 0.097 mg / m3, 0.049 m respectively for the optimal composition of chlorophyll a and Secchi-disk depth. Also, SVM model with the lowest input power also had the power to predict. The results of this study showed that the support vector machine has a high potential for predicting chlorophyll a and water clarity, and it can be used to determine the appropriate strategies for the management of the Choghakhor wetland.

Keywords: Choghakhor Wetland, Chlorophyll a, Root Mean Squared Error

Full-Text [PDF 829 kb] (1062 Downloads)

Type of Study: Research | Subject: Special
Published: 2020/06/30