In the Eye of a Very Mild Breeze


I'm sure some of you will have now read the 'news' item that was recently published in BBC Wildlife Magazine concerning estimates of great crested newt. The article appears to be an attempt to generate a story, where none really exists. Before explaining why I believe that the article is flawed I thought that it would be useful to provide a little background.

Back in 2006 I presented a talk at the Herpetofauna Workers Meeting entitled 'How Many Great Crested Newt Ponds are there in Kent?'. The summary for this talk is reproduced below:

"The great crested newt is believed to be experiencing a national population decline greater than that of the other widespread amphibian species. In Kent the species remains relatively widespread, displaying a county distribution that appears to be dependent upon pond density. Surveying for newts is labour intensive and requires multiple pond visits to reliably confirm presence/likely absence. Assessment of newt populations using simple counts is problematic due to variation in detection rate between ponds and observers. Habitat proxies are increasingly being used to predict the likely presence of great crested newt. This study collected habitat proxy data using a published Habitat Suitability Index. Ponds were ranked based on apparent quality and surveyed to determine the relationship between predicted and actual presence. "

The talk illustrated how occupancy data can be used in conjunction with information on the number of ponds to make predictions on how many of Kent's ponds are likely to be occupied by gcn. The process is really quite straightforward.

1. How is gcn occupancy influenced by pond quality?
2. How many ponds are there?



1. How is GCN Occupancy Influenced by Pond Quality?

The likely presence of great crested newt in a pond can be predicted using a published Habitat Suitability Index (HSI):

Oldham, R. S., Keeble, J., Swan, M. J. S. and Jeffcote, M. (2000) Evaluating the suitability of habitat for the great crested newt (Triturus cristatus). Herpetological Journal, 10, 143-155.

The HSI has previously been discussed on the blog and I have also prepared a summary of how the HSI can be applied elsewhere, but it is probably worth repeating some of the main points here.

The likely presence of great crested newt in a pond can be predicted by examining aquatic habitat features such as the presence of fish, waterfowl and water quality.

For ponds these data are used to calculate a HSI. The HSI is represented by a number from 0 to 1, the higher the number the more likely the pond is to be occupied by great crested newt. In order to help with the interpretation of a pond's HSI, ponds are classified with a description that is based on their apparent suitability for great crested newt. Thus:

It means a heck of a lot more to a landowner when you explain the likely presence of gcn in English rather than gobbledegook!

By undertaking detailed survey work in a large number of ponds across SE England I have found that ponds with relatively high HSI scores (good - excellent) frequently support great crested newt. Indeed, great crested newt have been recorded in more than 90% of the surveyed 'excellent' ponds! The relationship between HSI and actual presence is illustrated below:



Note also that although gcn are rarely encountered in ponds with very low HSI scores, they do sometimes occur.

By looking at a large number of ponds, I have also been able to estimate the proportion of ponds with each of the different quality ratings. For example, while 24% of ponds are likely to be 'poor', only 12% of ponds will be 'excellent'.

The overall % of ponds likely to support gcn will therefore be:

(available poor ponds x occupied poor ponds) +
.... +
(available excellent ponds x occupied excellent ponds)

For gcn I have generated an overall occupancy of 44%. This means that gcn will be present in almost half of the county's ponds! However, do consider that this is a county average. In some areas (e.g. countryside around Iwade), gcn occupy almost all suitable ponds. In other areas gcn may not be present at all, even in 'excellent' ponds. The presence of gcn in suitable ponds will largely be determined by the newt's local population status. A pond can be top notch quality, but if gcn are simply not present in the local area they may not have been able to colonise. Conversely, low quality ponds can be occupied, especially when they are situated close to another pond that already supports the species.

I also suggest that although a pond may be occupied by gcn, the amphibians may not necessarily breed there. My own survey work indicates that breeding success is more likely in ponds that are rated as 'good' or 'excellent'. The implications of this are that although gcn may occupy 44% of the county's ponds, they may only successfully breed in 27% of them.


2. How Many Ponds Are There?

In Kent, information on the number of ponds has been published by Kent County Council (KCC) and is freely available through the Kent Landscape Information System. From the website:

"The purpose of the K-LIS website is to enable better informed decision-making by providing detailed information on Kent's landscape and biodiversity. Primarily aimed at landowners, farm advisors and those involved in land use planning, the website also acts as a useful information source to members of the public. This site contains details on countryside access, landscape character, identifies opportunities for habitat creation and landscape restoration, the physical environment including soils and geology, the Kent habitat survey as well as areas designated for their conservation value."

The Kent and Medway Biological Records Centre have very kindly analysed this data on my behalf and suggested that there may be more than 41,000 ponds in Kent. Wow!

So, how many great crested newt ponds are there in Kent? Some simple maths:

41,000 ponds x 44% occupancy = 18,000 occupied ponds

That is an enormous number of occupied gcn ponds. However, as I very carefully explained during my talk at the Herpetofauna Workers Meeting, this is a preliminary finding and subject to verification. During the talk the audience were specifically told not to quote the figure due to concerns over the validity of the KCC pond count. What are those concerns? Look at the map below:



It is apparent that there are a large number of waterbodies on Romney Marshes, North Kent Marshes and North East Kent Marshes. These waterbodies are almost certainly ditches and drains - not ponds. Great crested newt can (and do!) breed in ditches and drains. However, the HSI was not developed for use on these waterbodies and does not take account of habitat features that may be unique to these linear waterbodies. For example, ditches often contain flowing water that could wash away gcn larvae. Ditches may also form an interconnected system that is seasonally flooded with brackish water and/or fish. The Kent pond data should more accurately be called Kent pond + ditches/drains data. At the present time it is unclear what impact this will have on the actual pond count. However, one thing is obvious: 41,000 ponds is an overestimate.


Some Conclusions

1. Kent is one of the best surveyed counties in the country and the amphibian survey data collected by KRAG volunteers is a tremendously useful resource.

2. Survey data has enabled me to estimate overall occupancy of ponds by great crested newt. As new data comes in estimates of occupancy may change a little. However, based on available information I predict that 44% of Kent's ponds are occupied by gcn.

3. GCN will not successfully breed in all occupied ponds. The loss of a single 'excellent' pond could therefore lead to the loss of gcn from several other lower quality ponds that are situated nearby.

4. Very few counties have detailed habitat data. Kent is extremely fortunate that KCC have funded a wide-ranging habitat survey. Data from this survey is currently used to inform conservation, management and planning across the county.

5. There are concerns over the accuracy of the Kent pond data. The quoted figure of 41,000 ponds is likely to be an overestimate. However, the actual number of ponds is considered to be in the same order of magnitude. Even if the inclusion of ditches and drains has doubled the current pond estimate, that still represents a tremendous number of ponds. Kent is probably one of the most import reservoirs of freshwater habitat in the country.

6. The KCC pond data may be inaccurate, but it is all we have. At this time I would prefer not to quote the figure of 18,000 occupied ponds for gcn in Kent. Yet this estimate remains valid based on the existing KCC dataset.

7. While any estimate of the numbers of a species are likely to have margins of error, I would urge KCC to consider reanalysing the Kent pond data so that we can make a revised and hopefully more accurate estimate of the number of gcn occupied ponds.



Next up, I will review the BBC Wildlife article and explain why I believe this particular journalist got things so wrong.


Lee Brady
KRAG Chairman

Posted: Wed - October 17, 2007 at 10:28 am        

©