Case Studies

The Challenge

European Protected Species (EPS) are plants and animals that are protected by law throughout the European Union. States are required under the Habitats Directive to protect the listed species and, in some cases, designate Special Areas of Conservation to safeguard their populations.

For example, the great crested newt (GCN) is an internationally important species which, following a dramatic decline in its populations across Europe, was given EPS protection. Due to the protected nature of this species, the Government has an obligation to give 6-yearly reports on its status.

There are a range of factors that may affect the long term distribution and abundance of the GCN populations, including, but not limited to, its range, habitat availability and population dynamics. In order to determine whether these areas are of a favourable status, data from field surveys and questionnaires need to be collected and analysed. However, given the widespread distribution of GCN in the UK, it is prohibitively expensive to conduct detailed field surveys on a 6-yearly basis. Therefore, the limited data available from various local and coordinated amphibian surveys need to be utilised.

The Solution

In the absence of a national, directed sampling plan, the GCN data available are ad hoc and vary considerably in their quality and coverage. Therefore, modelling approaches are required that account for sampling bias, for example, where greater survey effort being directed to areas where newts are more likely to occur in higher numbers has led to the data not being representative of the presence of the species nationally. Furthermore, as is common with most species data, only presence and not absence records are often available and presence records generally relate to GCN populations and not counts of individual newts.

Species distribution modelling (otherwise known as ecological niche factor modelling) can combine the patchy presence records along with environmental covariates, such as bioclimatic, land cover, elevation, and pond area and quality data, to predict the geographic distribution of species based upon their known distribution in environmental space.

Various species distribution models are available, but Maximum Entropy (MaxEnt) modelling is known to be particularly well-suited to noisy or sparse information, as is typical of species’ occurrence data, and has become increasingly popular in species distribution modelling in recent years.

The Value

MaxEnt modelling can provide estimates of the relative suitability of environmental conditions for GCN. Combining this with prevalence (occupancy rate) estimates, the probability of occurrence of GCN populations can be inferred. These results can be visualised, via choropleths (aka heat maps) for example, and, more importantly, can be used to aid local decision making on land-use planning and licensing, and targeting conservation and compensation mitigation measures

Though the MaxEnt modelling approach is able to provide useful estimates from the data available, it is important to recognise that any model can only be as good as the data that it is built upon. More extensive fieldwork programmes would be invaluable in helping to ensure that species distribution modelling can best aid public bodies in taking an evidence-based approach to monitoring and maintaining GCN populations.