Integrating Remote Sensing and Field Observations for Karst Morphological Analysis: Evidence from Rongkop, Gunung Kidul
Year: 2025
Type: conference-paper
Abstract: Karst landscapes in tropical regions, such as Gunung Kidul, Indonesia, are ecologically vital yet notoriously challenging to map at scale due to their complex morphology and widespread anthropogenic interference.While remote sensing offers scalable tools for identifying karst features, its accuracy remains questionable without rigorous ground validation-especially in areas where quarries, terraces, and vegetation can mask true geological signals.This study bridges this gap by integrating Sentinel-2 imagery and DEM-NAS with field-verified observations to evaluate the reliability of remotely detected karst morphologies in Rongkop Subdistrict.We employed a dualmethod approach: first, geomorphometric analysis (slope, curvature, TRI, doline density) derived from DEMs was used to classify landforms via decision tree classification; second, 142 landform units were validated in situ using GNSS surveys, lithological checks, and structured field protocols across 28 sites.Vegetation stress patterns (NDVI/EVI) and UAV-derived DSMs further refined the interpretation.Results show that while remote sensing successfully identifies cockpit-style isolated hills and multibasinal drainage (accuracy: 87.3%, κ = 0.82), over 30% of automated detections were false positives-primarily quarries misclassified as dolines.Field data revealed that karst topography is fundamentally controlled by lithological heterogeneity: resistant reef limestone (10%) forms elevated, vegetated residuals, whereas pervasive calcarenite (90%) facilitates subsurface dissolution and depression formation.A modified Karstification Index, calibrated against field dissolution rates, spatially correlates with observed morphological intensity.Our framework demonstrates that accurate karst mapping in data-scarce tropics requires not better algorithms-but deliberate, iterative dialogue between pixels and pebbles.This protocol offers a replicable and low-cost methodology for regional-scale karst assessment, directly supporting groundwater management and geoconservation planning.
Authors: Utama P P, Adha I, Dewi N K E S, Anggrasari H, Ajie M A, Zelvany Z T, Adelina J G, Ginting L M
Institution: Universitas Pembangunan Nasional Veteran Yogyakarta
Language: English
Subfield: Earth-Surface Processes
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDG: Life on land
Open Access status: hybrid
OpenAlex ID: W4415204446
FWCI:
Cites:
Cited by:
Related to: