Forest quality decline and restoration priorities in Indonesia’s New Capital Region

Abstract

Rapid development in the Capital City of Nusantara (IKN), East Kalimantan, Indonesia, is occurring within a landscape that still contains extensive humid tropical forest. This study quantifies spatiotemporal Land Cover change (2016-2025) and evaluates landscape-scale ecosystem condition using a Land Cover Quality Index (LCQI) derived from Sentinel-2 imagery, complemented by a DPSIR (Driving force–Pressure–State–Impact– Response) synthesis to frame restoration and management priorities. Land cover was mapped into 10 classes using a Random Forest classifier, achieving an overall accuracy of 85.2% (κ=0.82). Total forest cover decreased from 181,656 ha (70.8%) in 2016 to 138,700 ha (54.1%) in 2025 (-43,000 ha), including primary forest loss of 36,500 ha (-3,650 ha yr-1) and a marked increase in degraded forest (256 to 20,600 ha). Shrubland expanded by 32,600 ha, while built-up area increased from 8,400 to 27,200 ha. Consistent with these transitions, LCQI declined from 71.3 to 54.8, falling below the degradation threshold (≤65), a commonly used screening benchmark in environmental quality assessments. Hotspot analysis identified concentrated forest loss along infrastructure corridors in the northern and eastern parts of IKN, as well as near mining and river-valley conversion fronts. Insights from key-informant interviews (n=25) suggest that capital relocation, associated population growth, and investment flows are prominent drivers associated with land-conversion pressures. We do not directly measure biodiversity responses in the field; therefore, biodiversity implications are inferred from established relationships between forest loss/fragmentation and species viability. The driver–pressure linkages framed in the DPSIR framework should be interpreted as a structured synthesis of evidence and stakeholder perspectives rather than experimentally demonstrated causation. Accordingly, results are presented as spatial risk indicators to support restoration prioritization rather than as observed biological outcomes. Spatially explicit priorities include protecting remaining primary-forest blocks, rehabilitating degraded forest and shrubland-transition areas, and institutionalizing periodic LCQI-based monitoring to support the Forest City vision.