Denudation of Forested Mountainsides Detected by Satellite Observations

Abstract:

Extreme urbanization has exerted a high demand for materials such as soil, sand, and rock to satisfy building construction. Consequently, mining exploitation for building materials has led to denudation of forested mountainsides, inflicting serious deforestation with severe environmental impacts. Especially in tropical coastal regions, moisture-laden air mass forced up a windward mountainside forms orographic clouds and develops heavy precipitations. Heavy rains can generate intense runoff on steep and bare soil surfaces, without vegetation holding capacity to slow down the running water, and thereby cause flash floods, mud flows, or landslides, which may ironically surge toward and damage new buildings in the vicinity. To detect and monitor the denudation of forested mountainsides in the tropics, we have developed an initial algorithm using remote sensing data acquired by the Multi-Spectral Instrument (MSI) aboard Sentinel-2 satellites. Moreover, the algorithm accounts for complex landforms in mountainous areas based on the geomorphon concept (Jasiewicz and Stepinski, Geomorphology, doi:10.1016/j.geomorph.2012.11.005, 2013) to circumvent the ill-posed calculus of taking spatial derivatives of noisy topography data. In the algorithm implementation, we use Google Earth Engine to process time series satellite data, and employ the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) derived from satellite synthetic aperture radar (SAR) data to calculate geomorphons. To demonstrate the successful capability of this algorithm to detect denuded areas on forested mountainsides, we present the cases of coastal mountains in Tân Phước (Bà Rịa–Vũng Tàu Province) and in Nha Trang (Khánh Hòa Province), Vietnam, where mining activities have quarried extensive sections from the foothill to high elevations on the mountainous terrain.