Pieter's Blog

Welcome to Pieter Kat's official LionAid blog. Here you can follow Pieter's opinions, thoughts, insights and ideas on saving lions.

Lion population numbers in Africa - a new but once again flawed

In May this year, Jason Riggio, in fulfilment of the requirements of a Master’s Degree in Environmental Management at Duke University, NC, produced a thesis entitled “The African lion (Panthera leo leo): a continent-wide species distribution study and population analysis”. Riggio calls this “the most comprehensive assessment of the status and distribution of species to-date”. Further, he says that “the compilation of this information into a single database for future researchers and lion conservationists is a major outcome of this project”. I have reservations.


This document could perhaps begin to take on some misplaced importance, and therefore it is important to point out at an early stage the many inconsistencies and faults that led to some surprising conclusions by Riggio. Primary among these are the proposition that Africa’s lion population is much larger than some would conservatively estimate – well over 30,000 – perhaps as high as 35,000. Also, Riggio estimates that “more than half of the remaining lions in Africa reside in 11 viable populations contained within protected areas that have stable or increasing lion population trends (lion strongholds). Therefore lions are not currently threatened with extinction and it is unlikely that the total population of free-ranging lions in Africa will drop below 20,000 individuals. Given these findings, it is clear that new data based on field surveys are necessary to appropriately evaluate the legal status of the African lion”.


How did Riggio arrive at such conclusions? First, he refined the LCUs (Lion Conservation Units as designated by the IUCN in 2006) using four categories of data; land conversion, human population density, lion distribution from recent country-specific reports, and additional data from recent lion population surveys. Second, converted land (agriculture, etc) was excluded from possible lion habitats, as well as areas where there were more than 25 people based on GIS data. Third, lion distributions from past reports were accepted more or less uncritically, unless there had been a recent survey to show lions were no longer present in LCUs. Finally, “habitat patches” were designated, sometimes bigger than the designated LCUs based on human population density numbers, and then lion numbers extrapolated at times….


Still with me? Distilling all this down, Riggio basically uses existing data from “surveys” conducted by Bauer and Van der Merwe (2004 – 70% of data based on “guesstimates”) Chardonnet (2002 – never published in a scientific journal) IUCN (2006 – two meetings - western and central Africa, and eastern and southern Africa where delegates got together and proposed lion numbers for their countries - more guesswork). There are further reports by Chardonnet (2009) and Mesochina et al (2010) looking at specific countries. Again, nobody went out and counted lions, it was all done by interviews and questionnaires. Mesochina et al estimated that lions occur in 94% of Tanzania – at least Riggio was doubtful enough to exclude “occasional” observations.


Basically, one could propose that the current lion “survey” data is not much different from the following scenario: someone organizes a telephone or mail or personal interview scenario where 100 respondents (local experts since they live in the house) are asked how many mice they think they have. Some will say zero, some will say they are infested, hundreds at least. Take an average of the number of mice, so each house has about 50. Then extrapolate this across 200,000 houses in a city, and next thing you know there is an article in the newspaper that reports 10 million mice are swamping Birmingham. How many mice are there really? It takes a lot better data quality than that to make an informed decision.


Back to lions. Riggio then discusses many LCUs, and comes up with lion totals in each, usually prefaced with statements like “I propose…” “I suspect that…” “I propose a speculative population of …” “I tentatively retain…”. Nevertheless, except in cases where lions have been shown no longer to be present in identified LCUs by actual surveys, Riggio then rigorously accepts those former estimates in his analysis. And in some cases increases estimates when he judges the area of the identified LCU can be expanded over surrounding areas as long as there are fewer than 25 people/km2 living there and little satellite data indicating land conversion.


I have a bit of a problem with this. When Henshel and collaborators did a survey of 20 western African LCUs in 2010, they found 11 of them had no lions at all. They also found that many of these LCUs had semi-resident pastoralists and poachers. Land was not converted, and there were far fewer than 25 people per km2, yet the effect of those few people on the areas was significant and destructive to wildlife.


Next, Riggio uses the model of Bjorklund (2003) where that author proposes that a viable populations of lions need to consist of at least 50 prides in an area – and that 100 would be much better. Riggio makes a decision that a pride consists of 5 adult lions (1 male, 4 females? 2 males, 3 females?) and then decides that any area (by extrapolation, mainly) that contains about 250-500 lions is a viable population. This is a big mistake. The lion estimates delivered by respondents are a total population, not just adult lions. The total population consists of juveniles and subadults and potentially reproducing adults, and being very conservative, any lion population consists of at least half of the two former categories. So actually, you need a population of at least 500-1000 total lions to make up a viable population. Dave Youldon and I estimated there are no more than 5-6 lion populations that meet that target rather than the 11+ populations proposed by Riggio, and we will stick with that better informed estimate.


Finally, and this is big, Riggio proposes, based on his building layers and layers of supposition and extrapolation and models upon each other, that lions should not be considered in any way as needing specific attention from regulatory conservation organizations like the IUCN. Standing on his wobbly tower of creative interpretation, this pronouncement by a student who recently gained a Master’s degree is patent nonsense. He further proposes that IF protected areas (at least) get their acts together, there will be 25,000 lions in Africa in perpetuity. This is at best very hopeful, but sadly also encourages complacency at a time when this is least needed.






Posted by Pieter Kat at 10:14

No comments have been posted yet.

Add a new comment

Existing user

New user sign up