Condition Matters: Fine-Tuning Ecosystem Service Capacity with Ecosystem Condition

Ecosystem service (ES) capacity matrices are widely used for ES assessment and mapping due to their operational simplicity. However, standard matrices assign uniform capacity scores to ecosystem types without accounting for local variations in ecosystem condition, leading to systematic assessment errors. We present a methodological framework for integrating ecosystem condition into ES capacity matrices through ecosystem-specific condition thresholds. Ecosystem condition is characterized using two components: structural condition (e.g. vegetation cover, productivity, canopy height) and biological condition (e.g. biodiversity intactness, soil organic carbon, natural habitat proportion). We classify ecosystem services into five functional groups based on their dependence on these condition components. Based on this classification, we develop allocation tables linking condition states (each component is defined as low, moderatestandard, high based on condition indicatos) to adjusted capacity matrices. After testing different ecosystem-specific percentile thresholds for our test case area and our condition indicators we used the 20th-80th percentiles level to define condtion classes. Our results made clear that the condition values have to be assessed separately for each ecosystem type (urban, agricultural, forest and aquatic). The framework employs ecosystem-specific percentile thresholds (20th-80th) to define condition classes, avoiding systematic bias between ecosystem types while capturing within-type quality variation. Application to a 430-km² study area in the Hauts-de-France region (France) demonstrates that condition-adjusted ES capacity matrice assessments differ from unadjusted matrices by 0.58-0.81 units on average over the 0-5 ES scores scale (12-16% of the capacity range), with maximum local differences reaching 2.1 in a 0 to 5 scale units (42% of range). Validation against six independent ES reference datasets shows correlation improvements ranging from 0.035 to 0.132 (mean 0.078) when incorporating condition adjustments. The approach provides an operational method for implementing the SEEA EA conceptual linkage between ecosystem condition and service capacity, with applications in natural capital accounting, land-use planning, and environmental impact assessment.