Book Review

The Bird Atlas of Uganda
Margaret Carswell, Derek Pomeroy, Jack Reynolds & Herbert Tushabe, 2005. Oxford: British Ornithologists' Club & British Ornithologists' Union. 553 pp, many maps, figures and tables. ISBN 0-9522866-4-8. UK£55.

from ABC Bulletin 14.1 March 2007 pages 112 - 113.

The bulk of this book consists of maps and brief species accounts for all of the 822 species for which five or more acceptable records exist for Uganda. The maps are primarily those for Quarter Square Degrees (83 + another 14 with minimal land in the country), although if more detailed coordinates are available for any given record it is placed as accurately as possible, often resulting in several dots in each square or one which spans more than one square.

The book, however, commences with some typical introductory matter, including: Environment of
Uganda, which details landforms, climate, seasonality, major ecological zones and human impacts; a Brief History of Ornithology (and it is brief!); Overview of the Birds; Planning for Bird Conservation; and an introduction to the atlas itself. There follow c.450 pages of species accounts, some appendices, including additions to the Uganda list since January 2000, a list of poorly documented species, species with published records that are considered erroneous, a gazetteer, and rounded off with a comprehensive list of c.600 references, which is a goldmine in
itself.

Most of the introductory chapters provide essential background, but do not contain anything startling. However section 5 of Chapter 6 'Prediction of Habitat Suitability' is an interesting and innovative idea for atlases and despite some problems (fully admitted by the authors), for many species it works quite well and certainly helps to give a fuller picture of bird diversity than the actual records would do alone. From the start, the authors realised that coverage was likely to be patchy particularly in some parts of the country. The Kampala area and the national parks
of the west are quite well covered but there are often squares in the rest of the country, especially in the north, with few records or even none. So, in order to give a fuller picture of a
species' potential distribution rather than just relying on the actual records, the authors have determined where a species is likely to occur based on environmental variables. Full details are given of the process and its problems but very briefly, for those species with ten or more actual
records (excluding waterbirds) they use rainfall, altitude, vegetation and habitat, and human-based environmental variables for the points at which the species was actually recorded, to derive models that predict in which other areas the species might occur but was not necessarily recorded. However, inevitably with a relatively simple model, some anomalies result. The most obvious are some of the montane forest species whose predicted range extends into lowland
forests around Lake Victoria, and especially several species which are only known from Ruwenzori and the Albertine Rift are predicted to occur much further east, but there are quite lot of others. Overall, this technique was used for nearly 500 of the 822 mapped.

The maps are based largely on the Ugandan National Biodiversity Data Bank (UNDB) and include data until the end of 1999. Pre-1990 records are distinguished separately and derive largely from the many and various literature sources. However, no special field work was conducted and nor it appears were the many visitors to the country in recent years targeted. This is a shame because many of the maps do seem rather sparser of definite records than they should and perhaps could be. However, all records shown are fully documented in the UNDB and it does serve to highlight that all visitors should send their observations to the National Data Bank. Some visitors and tour
companies are excellent at this but many others do not bother (a comment which applies to many other countries too).

The species accounts accompanying the maps are usually brief (normally two per page). They consider taxonomy (the races which occur), habitat, and provenance of records (if this is important), whilst up to 50% of each account details any breeding records, though these are not distinguished on the maps. For many even quite common species there are none! As such, these accounts with the accompanying maps give an excellent summary of the current status of all
birds in Uganda. For species with fewer than five records there is a brief account but no map, and the appendices detail new records and dismiss erroneous ones. The total species list at the end of 2004 was 1,014 with five more still under review. This book will become the standard work on the birds of Uganda for many years and forms a major baseline. However, there are gaps and
there are clearly some large areas of the country which are virtually unexplored. Hopefully, if the political situation improves, these will be visited more frequently. Tour companies should be encouraged to visit these less well-covered parts and, of course, always to submit their records. After all, how can we conserve many of these birds unless we know where they occur?

Peter Lack

Last page update 28th May 2011