Spatial distribution of water quality in relation to phytoplankton abundance of Selat Tuba, Langkawi / Hasmida Muhamad

Phytoplankton is a microscopic marine alga that is mostly buoyant and floats in the upper part of the water body. Despite its size, it plays a vital role in sustaining the biological clock of marine life by serving as one of the primary providers of food and oxygen. However, their abundance greatly depends on the water quality of their habitat. The structure of the phytoplankton population changes in response to changes in water quality and other regulatory factors, which may or may not have an impact on the animal community and may result in a general loss of biodiversity, including a decline in fish production. Thus, this study aims to assess the water quality parameters (WQP) at Selat Tuba River by i) evaluating the spatial interpolation (IDW and Kriging models) for WQP estimation, ii) establishing a correlation between WQP and phytoplankton abundance, and iii) establishing regression equations for phytoplankton abundance. The WQPs involved are salinity, dissolved oxygen, Secchi disc depth, temperature (temp), and pH. The WQP and phytoplankton abundance were collected by in-situ sampling and assessed by laboratory experiments. The in-situ water samplings consist of 113 points and about 36 points for phytoplankton that were collected in December 2020 and December 2021. For the laboratory experiment, each individual phytoplankton was calculated in every 1 ml of water, and the total phytoplankton abundance was calculated in ind/mg3 units. To reduce uncertainty errors in the WQP, a Monte Carlo uncertainty analysis has been conducted. Spatial interpolation (IDW and Kriging models) was applied and tested; 70% and 30% of the water sampling data were used for model development and verification, respectively. The root means square error (RMSE) and mean absolute error (MAE) are applied to measure the goodness of fit of the regression equation for WQP estimation. The study found that pH and DO have the strongest correlation (0.948), while salinity and temperature have the lowest correlation (0.368). The IDW provides more robust models to estimate DO, pH, and temperature, with RMSE values of 0.517, 0.084, and 0.450, respectively. The optimal regression model for phytoplankton abundance estimation has an RMSE of 11.451 and an R2 of 0.3740. This study contributes to establishing the final regressions that can be employed to estimate water quality and phytoplankton abundance at the mangrove area as well as support the fishing industry's economy. Overall, the relationship between each WQP and phytoplankton had been established, the IDW interpolation model is proven to serve a more stable model compared to Kriging, and lastly the regressions for WQP and phytoplankton estimation has been established.