Impacts of land use and infrastructural changes on threatened Little Bustard Tetrax tetrax breeding populations: quantitative assessments using a recently developed spatially explicit dynamic modelling framework

Abstract

SummaryWith the combination of worldwide landscape changes and the uncertainty about the impact on species abundance and distribution, the value of spatio-temporal modelling tools is increasingly obvious. The Little Bustard Tetrax tetrax breeds on low-intensity arable cultivation and pastoral land and is currently threatened by diverse landscape modifications. The aim of this research was to predict Little Bustard population trends in the face of realistic scenarios of land use and infrastructure changes, applying a recently developed spatially explicit framework, based on the stochastic dynamic methodology (StDM). The application of this approach provided some basis to analyse the responses of breeding populations’ spatial distribution and abundance to the scenarios implemented. Since some of these scenarios represent local and/or regional risks to the viability of Little Bustard breeding populations, the results obtained demonstrate the potential of the proposed framework for landscape planning in the scope of the conservation of this threatened species. This approach also provides a promising baseline to support ecological risk assessments for other species, derived from ecological models with increased predictive power and intuitiveness to decision makers and environmental managers.

Changes in grassland management and linear infrastructures associated to the decline of an endangered bird population

“…(4) Finally, there was an overall increase of anthropogenic infrastructures (namely roads and power lines), which are known to negatively impact little bustards. Collisions with power lines are a major source of mortality for the species in the Iberia Peninsula 40 and power lines and roads are also a known source of displacement effects for the species [47][48][49][50][51][52] .…”

How good are the ecological assumptions and predictions made in the past? Insights from a dynamic modelling approach applied to changing landscapes

“…Our model falls into the category of the so-called hybrid models (Chen et al, 2011), i.e., modelling species distributions on the basis of large-scale holistic relationships (e.g., rural landscape dynamics) while at the same time considering the most important fine-scale processes (e.g., dependence on specific LULC). This is one of the major advantages of this type of models (Santos et al, 2013): combining different approaches and enabling information crossing at a local scale, allowing for a better understanding of rural landscape dynamics, and leading to emergence and complex system characteristics (Santos et al, 2013(Santos et al, , 2016a). If we consider that validation is a fundamental process to prove model applicability, our results demonstrate that the trends in the landscape composition and guilds richness were in general captured with credibility by the Santos and Cabral (2004) work.…”

“…That work proposed a novel spatio-temporal modelling methodology, the stochastic dynamic methodology (StDM), in order to predict relevant ecological trends in changing landscapes. The StDM is a hybrid modelling protocol combining statistical and dynamic modelling with geostatistical techniques to address complex spatially-explicit emergent problems, from the individual habitat patch to the whole landscape context (e.g., Santos et al, 2013Santos et al, , 2016a. While the parameters of the dynamic model within StDM can be calibrated directly from field knowledge of ecosystem characteristics and bibliographic information (ecologically driven) others, namely the holistic parameters, have to be estimated using statistical algorithms.…”

“…The statistical parameters, if sustained by a database that includes representative gradients, emerge from spatiotemporal ecosystem properties, might be used for simulation and prediction. This methodology minimizes drawbacks linked with model construction, such as parameterization and complexity, since part of the variables and parameters result from statistical estimation (Santos et al, 2013(Santos et al, , 2016a. Explicitly, the response variables emerging from the statistical analysis correspond to the core state variables under study while the explanatory variables are the pertinent environmental factors (ecological-driven).…”

“…Explicitly, the response variables emerging from the statistical analysis correspond to the core state variables under study while the explanatory variables are the pertinent environmental factors (ecological-driven). The StDM framework has been successfully tested in several types of ecosystems affected by gradients of change, namely in agroecosystems (Santos et al, 2013(Santos et al, , 2016a. Santos and Cabral (2004) also stated that the evaluation of the overall methodology and results, namely the simulated trends in Land use/Land cover (LULC) and the ecological indicators responses could only be achieved using chronosequential data.…”