Back

An improved habitat suitability index for the great crested newt

Seccombe, E.; Salguero-Gomez, R.

2022-12-19 ecology
10.1101/2022.12.15.520292 bioRxiv
Show abstract

Great crested newts (GCN) are a protected species whose conservation depends on the provision and protection of their breeding habitat. A habitat suitability index (HSI) developed in 2000 is extensively used in GCN conservation to assess breeding habitat quality. This two-decade old index is based on limited data and validation. Here, we introduce a new HSI with an improved ability to reflect GCN presence/absence in UK ponds. This proposed HSI is easier, requires less data, and predicts GCN presence/absence better than the previous index. To inform the new index, we used a dual approach to identify the relative importance of environmental criteria to predict GCN presence/absence. Firstly, we conducted a survey of 288 users of the GCN HSI to assess the perceived strengths and limitations of the existing index. Secondly, we analysed national datasets of GCN presence/absence and associated environmental data. Using these findings, we then tested various index modifications. The final modifications of the new HSI include: (i) using an arithmetic mean to combine variables, rather than a geometric mean, to reduce calculation errors and allow compensation between variables; (ii) excluding water quality and waterfowl impact as these lacked significant power to predict GCN presence/absence and were deemed inaccurate by HSI users; and (iii) changing the scoring relationship for pond area to better reflect current data and provide scores for ponds over 2000m2. Additionally, we present modified graphs to reduce errors in the calculation of pond density scores. We compared scores from the new and original HSIs using an independent dataset for validation, showing that the new HSI better reflects GCN present/absence (larger effect sizes and R-squared values) in comparison to the old HSI. Adopting this improved HSI will enable more effective conservation of the protected species via better-informed decision-making and monitoring.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.