A Model Web approach to modelling climate change in biomes of Important Bird Areas Jon Olav Skøien a, ⁎, Michael Schulz a , Gregoire Dubois a , Ian Fisher b , Mark Balman c , Ian May c , Éamonn Ó Tuama d a Land Resource Management Unit, Institute for Environment and Sustainability, Joint Research Centre of the European Commission, 21027 Ispra (VA), Italy b Royal Society for the Protection of Birds, RSPB, The Lodge, Potton Road, Sandy, Bedfordshire, SG19 2DL, UK c BirdLife International, Cambridge, Wellbrook Court, Girton Road, Cambridge, CB3 0NA, UK d Global Biodiversity Information Facility (GBIF), Copenhagen, Denmark abstract article info Available online xxxx Keywords: Web services Ecological forecasting Model Web Bird conservation Protected Areas (PA) are designated to conserve species and habitats and protect against anthropogenic pres- sures. Park boundaries, however, offer no protection against climatic change and where boundaries are actual constructions, they may also act as physical barriers to species movements to new suitable habitats. The means for assessing the consequences of climate change on ecosystems and for identifying gaps in PA connectiv- ity are therefore a conservation priority. The complexity of the scientific questions raised requires a multi- disciplinary approach given the variety of the information required. This includes species observations and their theoretical distributions, as well as ecosystem assessments and climate change models. Such complex ques- tions can be more easily dealt with if there is appropriate access to data and models, a strategy endorsed by GEO-BON, the Group on Earth Observations Biodiversity Observation Network. In this paper, we show how data and models recently made available on the World Wide Web can be coupled through interoperable services and used for climate change forecasting in the context of Important Bird Areas (IBAs) and how, for any bird spe- cies described in the databases, areas can be identified where the species may find a more suitable environment in the future. As presented, this is an example of the Model Web. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Species distribution models (SDMs) are typically used for identify- ing the suitability of habitats based on observations of the species and a set of environmental indicators assumed to include the species' niche. There are a range of such models used in ecology (Guisan and Zimmerman, 2000). A relatively widely used method is based on the Mahalanobis distance to create environmental suitability maps (ESM) (Clark et al., 1993; Knick and Dyer, 1997; Rotenberry et al., 2002). Another method is the MaxEnt method (Phillips et al., 2006) which is based on the creation of pseudo-absence locations. Common to these and other models is that they can be used to gen- erate suitability or similarity maps which can help in identifying regions where a given species is more likely to be observed. These models can therefore be also used to define those locations where species could migrate should their current habitats become unsuitable as a result of human activities. In addition to modelling the potential distribution of species, such models can also be used for estimating similarities between ecosystems found in protected areas and those observed in their surroundings. This approach uses the size of the area with a similarity level above a certain threshold as a measure of the ecological uniqueness of a national park (Hartley et al., 2007; Nelson et al., 2009). There are a range of challenges in using species distribution model- ling for predicting the effects of climate change (Sinclair et al., 2010). In this paper, we will focus on how to simplify the use of such models through coupling of interoperable web services, a concept that is also referred to as the Model Web (Geller and Turner, 2007). In the Model Web, a simple interoperable web service which can be either offering or processing data, and delivering the result in a standardized format, can be reused by the next service. While many web services are already available, only a few of these are sufficiently interoperable for the tasks outlined in this paper. In the ecological world, many services are Web Map Services, where maps can be overlaid in a web client. These services are, however, usually deployed for improving the visualisation of the environmental context of the analysed information rather than for delivering new products which can be further processed. It is the purpose of this paper to discuss how a number of interoper- able web services can interact with each other and produce new infor- mation. The services described have been developed mainly in relation to establishing the Digital Observatory for Protected Areas (DOPA), a biodiversity information system (Dubois et al., 2010a, 2010b) devel- oped by the Joint Research Centre of the European Commission in Ecological Informatics xxx (2012) xxx–xxx ⁎ Corresponding author. E-mail address: jon.skoien@jrc.ec.europa.eu (J.O. Skøien). ECOINF-00363; No of Pages 6 1574-9541/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ecoinf.2012.12.003 Contents lists available at SciVerse ScienceDirect Ecological Informatics journal homepage: www.elsevier.com/locate/ecolinf Please cite this article as: Skøien, J.O., et al., A Model Web approach to modelling climate change in biomes of Important Bird Areas, Ecological Informatics (2012), http://dx.doi.org/10.1016/j.ecoinf.2012.12.003