The much-maligned wild elephant census in Sri Lanka

Friday, 16 September 2011 00:55 -     - {{hitsCtrl.values.hits}}

I do not think any recent issue related to nature and environment has generated such an avalanche of news, views, controversies, and information, as did the now concluded wild elephant census of the Department of Wild Life Conservation (DWLC) of Sri Lanka.

The DWLC study covered 65,000 sq km (the total area of Sri Lanka) in three days, utilising farmers, service personnel and a few experienced DWLC field officers and rangers, and came up with a figure of exactly 5,879 elephants in Sri Lanka!

In spite of an unanimous call by the majority of the world-renowned scientists and elephant researchers, both local and foreign, that the methodology used for the DWLC census was seriously flawed, the programme went ahead and, in a short space of two weeks, some detailed findings have been released.

I also delved into this issue at the very outset, writing several articles about the issues that were prevalent. I do not propose to get down to another treatise on the subject, for the simple reason that, like all previous attempts, it will be of no use – and, in any event, the ‘deed is now done’.

So what I have done is frame eight questions to the DWLC regarding this matter, as follows:

1. Was this project a ‘census’ or a ‘population estimate’?

There appears to be some confusion about the very focus and objective of the programme. Some officials at the DWLC have conceded to me (off the record) that this was only ‘an rough estimation of the population,’ while some of the experts who have been involved in the project say this is was to study the population structure and that the count was actually a by-product.

It is needless to say that if the overall objective of any project is ambiguous, then all the strategy, plans and methodology that follows will also be flawed.

2. Given that the methodology applied was exclusively ‘waterhole-based’ how accurate was the count (if it was a count)?

In utilising a ‘quick and dirty’ method such as a waterhole study, that too, over a short period of 72 hours, what is the accuracy of the estimate arrived at? Surely anyone knows that using a quick and easy method for evaluating something will yield approximate results only.

I am not in any way debating the methodology at this late stage. There are times we all have to use an approximation to arrive at some result. All I am saying is that we must know what the error margins are! How can we say that there are today 5,879 elephants, down to the last integer, when a very broad based approximate method was utilised for the ‘census? I would have been happier with ‘approximately 5,500-6,000 elephants,’ perhaps.

3. How were the 987 males identified?

Anyone who understands elephants in the wild will agree that distinguishing between males and females is the one of the most difficult things to do. Even seasoned experts sometimes make mistakes, because unlike in most other animals, the elephant’s external genital differences are not at all prevalent and obvious.

So how did some 3,000-odd, hurriedly-trained ‘farmers, service personnel and villagers’ undertake this feat in a short time span of 72 hours? I have very serious doubts about the accuracy of this information.

4. How many waterholes were covered?

A cursory glance at an irrigation/waterways map of Sri Lanka will reveal that the island, especially the north, north east, north western and eastern areas, are pockmarked with an array of small waterways and tanks. This is to be expected, since Sri Lanka once boasted of some of the most advanced and complex irrigation networks in the world at that period of time.


It is a well-known fact that these areas are the main wild elephant habitat. We also know from reliable scientific study that some 60-70% of all wild elephants live outside national parks, in these very same areas. So I am at a loss to comprehend how all these tanks and water ways could have been ‘covered’ in just three days. I believe that there would have been a very large error margin, on the negative side (i.e. undercounting).

5. Can you confidently say that there was no double counting?

Elephants are known to travel great distances in short space of time, if and when they decide to do so. A single animal could travel up to 50-60 km in a day while herds may move some 30 km or so. Therefore, there is a very real possibility that some elephants who were counted at one waterhole could be seen in another water hole in the vicinity a day or two later.

Unless there is some distinguishing feature of a single animal (like a broken off tusk, or deformity) it is very difficult to identify individuals to the untrained eye. So I am of the opinion that there could be a large error margin, this time on the positive side (i.e. over counting).

6. How can you say that the population is healthy?

The state of the heath of an elephant population or herd is judged not by the number of calves alone. (In this case supposedly some 1,100 calves have been recorded, which is about 18% of the total 5,879 counted.

A healthy elephant population should have a good balanced ratio of adult bull elephants and mature cows, a reasonable amount of sub-adults, as well as adolescents, in addition to calves. A preponderance of calves only may actually indicate a more serious situation, where although reproduction may be taking place, the calves may be dying off before they grow up.

7. If the count in the south of 1,086 elephants includes Yala and Uda Walawe National Parks, there are some serious errors in this data

There seems to be no other area they can be categorised into in the DWLC analysis – and no clarification could be obtained from the DWLC when writing this article.

It is interesting to note that according to the information released by the DWLC, the south records 1,086 elephants, where there are two national parks well known for elephants.

At Uda Walawe National Park (UWNP) there is currently today, one of the longest ongoing research programmes, which is studying the population dynamics, social behaviour and vocal communication of elephants.

In this study, led by Dr. Shermin de Silva of the University of Pennsylvania, with the affiliation of one of the foremost elephant experts in the country Dr. Devaka Weerakoon, it has been revealed that the UWNP alone is carrying between 800-1,160 elephants, both transient and resident. This is the data based on a laborious five-year-long study, by some world-renowned scientists at UWNP (

It is also pertinent to note that after five years of studying the elephants, day in and day out, in a confined space of approximately 350 sq km (the area of the UWNP), these scientists are still cautious of giving exact numbers and quote ‘800-1,160’.

The DWLC study, on the other hand, covered 65,000 sq km (the total area of Sri Lanka) in three days, utilising farmers, service personnel and a few experienced DWLC field officers and rangers, and came up with a figure of exactly 5,879 elephants in Sri Lanka!

8. How were the two elephants said to be in the Sinharaja region counted?

There definitely are a few elephants left in the Sinharaja region, and I myself have seen two of them, in the buffer zone on the Deniyaya side. But there are no ‘waterholes’ in this area! I actually have no issues with this count of two, but the point I am trying to make is that there is a serious confusion and ambiguity in this whole process.

My reading is that this estimate of two elephants in the Sinharaja area was based on hearsay and ‘guestimates’ and were actually not sighted or counted.

(The writer is Project Director, SWITCH ASIA Programme – Greening Sri Lankan Hotels, CCC Solutions and a former President of the Tourist Hotels Association of Sri Lanka.)

Recent columns