How to predict plant functional types using imaging spectroscopy: linking vegetation community traits, plant functional types and spectral response

摘要: 1.The comparable and integrated nature of plant functional types advances in high spectral resolution remote sensing techniques (i.e., imaging spectroscopy) make their combination highly interesting for spatially continuous repeatable large-scale ecosystem monitoring. Depending on physical environment stress, plants invest co-varying biochemical structural traits, influencing characteristics vegetation. These traits are assumed to bear a more direct causal relationship than life/growth forms. However, the connection between vegetation community's response remains be established. 2.We assessed correlation structure i) (biomass, dry matter content, nitrogen neutral detergent fibre content), ii) forms, iii) seven two categories (strategy types, indicator values) collected heterogeneous alpine grassland. We then used airborne spectroscopy data from same area model predict forms at community level using partial least squares regression (PLSR), validated our models based an independent dataset. 3.We found correlations many tested. Using data, we successfully modelled predicted most (R2 max. = 0.56) all 0.62). performance decreased substantially during external validation overall consistency was low (average change R2 72%), while type were much consistent 20%). Based findings, developed conceptual framework theory methodology ecology link its signature. 4.Our results encourage use order aid monitoring ecosystems, which is particularly important given increased availability prospective launches spaceborne instruments near future. This article protected by copyright. All rights reserved.