Sensing Total Suspended Solid in a Large Artificial Lake Using Landsat-OLI: A Case Study in Gajah Mungkur Reservoir, Indonesia

Abstract

Total Suspended Solids (TSS) concentration is one of the key indicators of water quality, reflecting the health of reservoir and watershed ecosystems. Accurate and timely monitoring of TSS is essential for effective management and environmental protection. However, existing models frequently fall short of providing reliable TSS estimates. To address this gap, this study evaluates ten current TSS estimation models and develops a novel model based on Landsat-8/9 OLI data for the Gajah Mungkur Reservoir. The objectives include generating a monthly TSS map and assessing water quality changes over time. We utilized Landsat-8 Level-2 data acquired on March 9, 2023, corresponding with in situ TSS observations collected within the reservoir. The algorithm for TSS estimation was developed through a systematic exploration of various combinations of band ratios derived from bands 2 (Blue), 3 (Green), 4 (Red), and 5 (Near Infrared). Our evaluation revealed that the ten existing models exhibited low accuracy in estimating TSS concentrations in Gajah Mungkur Reservoir, highlighting the need for an improved algorithm. In response, we applied bootstrapping techniques and developed a novel model that utilizes two-band ratios (NIR/Green and Red/Green) of Landsat-8/9 Level-2 data to estimate TSS more accurately. The new model has an R² of 0.78 and a root mean square error (RMSE) of 2.44 mg/L. Hence, we applied the new model to eleven Landsat 8/9 images spanning from February 2022 to March 2023 effectively capturing spatial and temporal variations in TSS concentrations in the Gajah Mungkur Reservoir. The model not only provides accurate TSS estimates but also captures the distribution patterns over time, offering valuable insights for environmental monitoring and management.