Modelling of total suspended particulates in Malaysian coastal waters using remote sensing techniques

This study focused on environmental remote sensing with the objective of constructing a remote sensing algorithm to determine Total Suspended Particulate (TSP) concentrations in Malaysian coastal surface waters. Other objectives included coral reef mapping and production of quantitative map of [TSP] using the remote sensing algorithm at the study area which was Tanjung Rhu, located northeast of Pulau Langkawi, Peninsular Malaysia. Measured [TSP] varied from 93.92 ± 50.10 mg/L to 148.65 ± 45.39 mg/L. The biogeographic distribution of the reef in Tanjung Rhu was mapped and the hermatypic coral species was identified taxonomically. Results were compared to a control site, Teluk Datai located northwest of Pulau Langkawi. There were 37 coral species in Tanjung Rhu and 76 species in Teluk Datai. The Jaccard's score was 27% indicating that the two reefs were quite diverse in their coral compositions. The development of a remote sensing algorithm is deemed necessary to provide a synoptic view of the potential problem within the coastal waters due to the early coastal development in Tanjung Rhu. Sedimentation studies showed sediment fractions were mainly biogenic materials and lithified sediments. Sediment fallout rates in Tanjung Rhu were 1,403.48 ± 125.60 g/m2/day (dry season) and 6,550.77 ± 641.43 g/m2/day (wet season). In Teluk Datai the sediment fallout rates were 1,532.99 ± 201.81 g/m2/day (dry season) and 12,446.45 ± 237.81 g/m2/day (wet season). The remote sensing algorithm, Suspended Particulate Algorithm for Coastal Remote Sensing (SPACoRS) developed from a modified Simple Radiative Transfer Model is defined as [TSP] (mg/L) = 0.6668e4 .3892x , where x represents (Rrs-toaTM31RrstoaTM2) ratio. SPACoRS is designed to determine [TSP] of 30 - 275 mg/L with primary material of TSP having high backscattering and low absorbance values. Sensitivity of SP ACoRS to produce higher accuracies was limited to TSP ~ 15Omg/L. SP ACoRS' s accuracy using Landsat Thematic Mapper data was 66%.