Satellite Assessment of Forest Health in Drought Conditions: A Novel Approach Combining Defoliation and Discolouration

Abstract

During the summer of 2022, heat waves exacerbated drought conditions across Europe, significantly deteriorating Slovakia’s forest health (FH). The main symptoms were defoliation and discolouration (mainly browning). According to the literature, completely brown leaves/needles are considered defoliation, and premature yellowing halts assimilation and reduces production. Thus, evaluating FH based solely on defoliation may underestimate the impact severity. To address this issue, we proposed a formula that integrates both defoliation and discolouration metrics. Then, by linking terrestrial and satellite data (a mosaic from Sentinel-2 and Landsat 9), regression models were developed using two-phase sampling to estimate defoliation, discolouration, and their combination. In the first phase, the Gram–Schmidt transformation of four satellite mosaic bands was used to derive two orthogonal components: one optimized for FH estimation (NSC2) and one for eliminating the influence of species composition on FH classification (NSC1). In the second phase, ground data were collected for the construction of a regression and to improve the first-phase results. The NSC2 component showed a strong correlation with defoliation, discolouration, and their combination. The standard error of the estimate was ±9.7% and the R2 was 0.83 for the combined symptoms, which enabled a detailed assessment of the intensity of forest damage. Our method is independent of tree species and has potential in FH assessments of temperate forests in Europe.