Mapping plant strategy types using remote sensing

Aims: The three-strategy (CSR) model proposed by Grime constitutes one of the most established systems for plant functional types. The primary strategies (competitive ability, adaptation to severe stress and adaptation to disturbance) relate to the productivity and level of disturbance at a given site. Accordingly, their change in space and time may serve as an illustration and measure of key processes such as succession, eutrophication and habitat shift. Here, we make use of the known links between the three primary strategies to traits with potential relevance for canopy reflectance, and test whether remote sensing is able to reproduce the spatial pattern of strategy types. Location: A raised bog and minerotrophic fen complex, Murnauer Moos, Germany. Methods: Field data on the distribution of plant strategies in sample plots were regressed against canopy reflectance using partial least squares regression. The resulting models were validated and applied to airborne hyperspectral imagery on a per pixel basis. The resulting local maps for each strategy type and their combined representation in an RGB colour composite were interpreted in terms of plant species composition and environmental constraints. Results: All three primary strategy types could be mapped using remote sensing. Reflectance spectra related to competitive ability and adaptation to severe stress suggest that typical traits linked to these strategies exerted a direct influence. On the other hand, species with low cover values played a decisive role for the strength of the statistical relationship between reflectance and strategies. Because these species have a low impact on canopy reflectance, their contribution is better explained by their role as proxies for covarying variables such as the total cover of dead plantmaterial. Conclusions: Our study demonstrates the potential to detect community strategy type composition using hyperspectral remote sensing, providing direct insights into spatial ecological patterns. By illustrating the exposure to stress, competition and disturbance, the derived maps of functional traits are potentially useful for applications in nature management and for the monitoring of functional shifts in ecosystems. As a next step, they can be easily combined into maps of functional diversity. Upcoming satellites with higher spectral resolution will improve access to this kind of spatial information.