EUVIMEDEUVIMED entdecken
Crossref

TINJAUAN INTEGRASI MACHINE LEARNING/DEEP LEARNING DAN DATA SPASIAL RESOLUSI TINGGI - SANGAT TINGGI UNTUK PENLIAIAN JASA EKOSISTEM AKIBAT PERUBAHAN IKLIM

Laode Rustam

Jurnal Celebica : Jurnal Kehutanan Indonesia · 10.07.2026 · 7 (1) · 108-123

DOI: 10.65078/jc.v7i1.237 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Jurnal Celebica : Jurnal Kehutanan Indonesia, online erschienen am 10.07.2026. Verfasst von Laode Rustam.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Global climate change and anthropogenic pressures induce massive degradation on the stability of Ecosystem Services (ES). The inherent limitations of conventional remote sensing in resolving mixed pixel biases across heterogeneous landscapes demand methodological innovations. This Systematic Literature Review (SLR) aims to synthesize global literature regarding the efficacy of integrating Machine Learning (ML) and Deep Learning (DL) algorithms with high to very high resolution (HR-VHR) spatial data for precise ES assessment. Through a structured analysis of reputable journal articles, the synthesis reveals that multi-scale spatial data fusion—combining satellite imagery (Sentinel-2/Landsat) and airborne platforms (UAV/Drone/LiDAR) significantly eliminates detection bottlenecks for pure spectral micro-objects. Advanced intelligent algorithms, such as Object-Based Image Analysis (OBIA) and multi-task convolutional networks, successfully formulate a functional scaling bridge. This bridge effectively reconciles local-scale aboveground biophysical samples with continental-scale subsurface soil physical and biological characteristics into continuous spatial visualizations of 30–90 M raster grids. The synthesized findings confirm a substantial increase in quantification accuracy above 80% for regulating services (pollination and greenhouse gas mitigation), provisioning services (forest biomass), and supporting services (topsoil microbial biogeography and wetland habitats). However, model dependency on site-specific training data and susceptibility to spectral noise caused by extreme weather anomalies remain critical challenges. This review concludes that AI-driven multi-sensor synergy provides a robust and sustainable predictive baseline, which is paramount for policymakers in formulating Nature-Based Solutions to mitigate global climate change impacts.

Zitieren

APA

Rustam, L. (2026). TINJAUAN INTEGRASI MACHINE LEARNING/DEEP LEARNING DAN DATA SPASIAL RESOLUSI TINGGI - SANGAT TINGGI UNTUK PENLIAIAN JASA EKOSISTEM AKIBAT PERUBAHAN IKLIM. Jurnal Celebica : Jurnal Kehutanan Indonesia, 7(1), 108-123. https://doi.org/10.65078/jc.v7i1.237

Vancouver

Rustam L. TINJAUAN INTEGRASI MACHINE LEARNING/DEEP LEARNING DAN DATA SPASIAL RESOLUSI TINGGI - SANGAT TINGGI UNTUK PENLIAIAN JASA EKOSISTEM AKIBAT PERUBAHAN IKLIM. Jurnal Celebica : Jurnal Kehutanan Indonesia. 2026;7(1):108-123. doi:10.65078/jc.v7i1.237

BibTeX
@article{Rustam2026,
  author = {Rustam, Laode},
  title = {TINJAUAN INTEGRASI MACHINE LEARNING/DEEP LEARNING DAN DATA SPASIAL RESOLUSI TINGGI - SANGAT TINGGI UNTUK PENLIAIAN JASA EKOSISTEM AKIBAT PERUBAHAN IKLIM},
  journal = {Jurnal Celebica : Jurnal Kehutanan Indonesia},
  year = {2026},
  volume = {7},
  number = {1},
  pages = {108--123},
  doi = {10.65078/jc.v7i1.237}
}
Kostenlos starten →