What's happening Wednesday

What’s happening Wednesday – Using species models for conservation: the value and the pitfalls

The Malay Civet

Many new computer tools are being used to help inform conservation actions on the ground. Of particular interest is where different species are found, what forest types they live in and using this information to produce distribution maps.
Recently a large group of researchers, including OuTrop’s Dr Susan Cheyne, decided to test the accuracy of a model called Maximum Entropy or MaxEnt. This programme uses records of animal presence and associated habitat and/or weather data to predict where else a particular species should be found.As with any model, the results are only as good as the data you put in. So when considering the whole island of Borneo, if only a few sightings of a species are included the model may greatly over- or under-predict possible distribution.
In the recent paper entitled “Theimportance of correcting for sampling bias in MaxEnt species distributionmodelshttp://onlinelibrary.wiley.com/doi/10.1111/ddi.12096/abstractthe researchers decided to use a common species, the Malay civet Viverra tangalunga, to test the accuracy of this modelling. The model used 291 records of the Malay civet from 2001 to 2011 together with 25 environmental input variables. These records included sightings and camera trap data.

Map showing the locations of all records used in the paper.

The majority of the records were from Malaysian Borneo and only OuTrop had any presence data for the southern part of Borneo. This increased sampling effort in Malaysian Borneo could lead to a possible sampling bias in data. While this could lead to inaccurate predictions of species presence, the research team found that a substantial improvement in the quality of model predictions can be achieved if uneven sampling effort is taken into account, thereby improving the usefulness of species conservation planning.