CD Skripsi
Analisis Indeks Kekeringan Meteorologis Lahan Gambut Di Pulau Bengkalis Menggunakan Google Earth Engine (Gee)
ABSTRACT
This study analyzes meteorological drought on peatlands in Bengkalis Island using
the Standardized Precipitation Index (SPI). Rainfall data from the Global
Precipitation Measurement (GPM) satellite for the 2015–2024 period were
processed through two approaches: manual calculation based on the Gamma
distribution using Microsoft Excel and automated computation with Google Earth
Engine (GEE). Validation was conducted by comparing SPI values between the two
methods using Percent Agreement and Mean Absolute Error (MAE). The results
show that the SPI values exhibit a consistent pattern, with an average classification
agreement of 89% and a low MAE (0.15–0.16), confirming that GEE is reliable and
more efficient than the manual method. The relationship between SPI and fire
hotspots from FIRMS data was examined in three grids (6, 8, and 9). Findings
indicate that hotspots generally increase when SPI is negative (dry conditions),
particularly below −1.0, although hotspots also appear under normal or wet
conditions due to anthropogenic factors. This study highlights that meteorological
drought significantly contributes to peatland fire risk, but integrated monitoring of
SPI and hotspot data is crucial as a basis for early warning systems and fire
mitigation in Bengkalis Island.
Keywords: meteorological drought, SPI, Google Earth Engine, fire hotspot,
peatland
Tidak tersedia versi lain