TIME SERIES ANALYSIS FOR MONITORING SEAGRASS HABITAT AND ENVIRONMENT IN BUSUANGA, PHILIPPINES USING GOOGLE EARTH ENGINE
Keywords: google earth engine, seagrass, water quality, time series, Landsat, MODIS, spectral unmixing
Abstract. Seagrasses are marine flowering plants which are part of a highly productive coastal ecosystem and play key roles in the coastal processes. Unfortunately, they are declining in area coverage globally, and seagrass losses can be attributed to climate change such as sea-level rise, increase in sea surface temperature, and decrease in salinity, as well as human-related activities. The objective of this research is to assess the historical changes in the seagrass habitat and environment of Busuanga, Philippines using time series data available in the Google Earth Engine (GEE) platform. These include satellite data such as MODIS, Landsat 5, 7, and 8, and SeaWIFS. Reanalysis data such as HYCOM was also utilized in this research. Results from HYCOM data show that there has been a 0.0098 °C increase in the sea surface temperature per decade in Busuanga while MODIS data indicates an increase of 0.0045 °C per decade. Moreover, HYCOM data also shows an overall average of 0.76 mm in sea surface elevation anomaly and a decreasing trend in salinity values at 0.0026 psu per decade. Chlorophyll-a concentration has a minimal increase based on results from MODIS and SeaWIFS. Aside from changes in water parameters, changes in the land also affect seagrasses. Forest loss may cause increased siltation in the coastal ecosystem which can lead to seagrass loss. Based on the results of Landsat satellite image processing, there has been forest cover loss in Busuanga with the highest loss occurring in 2013 when super typhoon Yolanda ravaged the island. Lastly, results from the linear spectral unmixing of 778 Landsat images from 1987–2000 show that the average percent cover of seagrasses in Busuanga were declining through the years.