Abstract

Semi-natural grassland habitats have markedly declined from their historical coverage, thus causing substantial losses for agricultural biodiversity and establishing a consequent need to spot the remaining habitat patches. These patches are generally remnants of once larger habitat areas, formed by uninterrupted and low-intensity management for centuries, but then later being isolated and fragmented into smaller pieces. In the light of this development, past landscape phases have a crucial role for the present existence of semi-natural grasslands. The importance of historical factors has been indicated in many studies but evaluation of their added value, or actual site-specific effects compared to observations of only the present landscape characteristics, is not generally provided. As data related to the past is often difficult to obtain, tedious to process and challenging to interpret, assessment of its advantages and related effects – or consequences of potential exclusion – would be needed.In this study, we used maximum entropy approach to model the distribution of Fumewort (Corydalis solida) which in the study area is a good indicator of valuable semi-natural habitats. We constructed three different models – one based on only the contemporary environment with expected indicators of habitat stability, one solely on the historical landscape phases and long-term dynamics, and one combining variables from the past and the present. Predictions of the three models were validated and compared with each other, followed by an analysis indicating the similarity of model results with known Fumewort occurrences. Our results indicate that present landscapes may provide workable surrogates to delineate larger core habitats, but utilization of historical data markedly improves the detection of small outlying patches. These conclusions emphasize the importance of previous landscape phases particularly in detecting marginal semi-natural grassland habitats, existing in contemporarily suboptimal conditions and being prone to disappear if no further actions are taken.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call