A fascination with (lesser) flamingos
18 March, 2025
By David Harper, Honorary FBA Research Fellow; Emeritus Professor, University of Leicester, david.m.harper@icloud.com
David Harper led research teams to Kenyan and Tanzanian lakes from 1982 until his retirement in 2015. In this – his second of two recent articles – David recalls his work with soda lakes and lesser flamingos.
Edited by Rachel Stubbington, Nottingham Trent University
Rachel is both a Fellow of the Freshwater Biological Association and long-standing Editor of FBA articles. If you would like to submit an article for consideration for publication, please contact Rachel at: rachel.stubbington@ntu.ac.uk
Lesser flamingos (Phoeniconaias minor) are fascinating birds (Figure 1). The only vertebrate to subsist on bacteria and algae, they mostly filter-feed the colonial cyanobacterium spirulina (Arthrospira fusiformis) from the surface waters of tropical African soda lakes. They also scrape cyanobacteria and algae from mud and rocks in shallow littoral and intertidal areas.
They are one of six flamingo species, and one of two in the Old World. They have four populations; the largest, of about 1.5 million birds, is in East Africa, southern Africa and India hold about 0.5 million, and West Africa perhaps 20,000. Each isolated population only breeds in a remote, predator-free location.
Figure 1. An adult lesser flamingo.
Introduction – how flamingos?
My first academic job, in the late 1970s, was as Science Coordinator in the Adult Education Department at Leicester, which involved teaching field ecology to local naturalists. With the collaboration of Nairobi University, I organised a field course in Lake Naivasha; we arrived in December 1982 and worked in partnership with their students for six weeks. We returned in 1985 and 1989 – which exhausted the limited pool of Adult Education students. I then applied to the US Earthwatch Institute, a charity doing the same thing: recruiting what we now call citizen scientists to assist with research in attractive locations! They funded my research for the next 18 years.
By 1999, we had collected enough data to organise an international scientific conference with local and international limnologists who were active in Kenya. William Kimosop, Warden of Lake Bogoria National Reserve, applied to speak. His talk comprised beautiful slides of the lake and its inhabitants, but he said nothing was known about its science. Suspecting my work on Lake Naivasha was complete, I applied to Earthwatch to move some of my research camps to Lake Bogoria, and we began in 2000.
Serendipity – why flamingos?
We had planned to study the limnology of Bogoria – a little-known, deeper (max. ~20 m) soda lake – to compare it with the better-studied, shallower (max. ~4 m) Nakuru. We arrived at the lake shore expecting solitude, but found >50 journalists photographing the carcasses of thousands of lesser flamingos. Hundreds were dying every day and being scavenged by marabou storks, eagles, baboons, hyenas and warthogs (Figure 2). We rethought our plans and diverted effort into studying the death rate and carcasses, as well as limnology and biodiversity (Harper et al. 2003).
Figure 2. A dead lesser flamingo scavenged by a marabou stork (left); explaining to international journalists what I thought was happening (centre) in a population of over a million birds at Lake Bogoria (right).
The mortality had subsided during our 2001 to 2005 visits, but a few individuals were still dying each time we returned. I persuaded wildlife veterinarian Dr J. Lindsay Oaks, who had studied the deaths of Gyps vultures in India and Pakistan, to come and study lesser flamingos. He autopsied fresh carcasses and initially confirmed the main cause of death as Mycobacterium avium: tuberculosis (Oaks et al. 2006). Sadly, Lindsay died before publishing his full data, but no other study had convincingly identified other causes. The publicity over the mortality event in the year 2000 had generated scientific interest and new investigations had suggested other causes: heavy metal pollution, cyanobacterial toxins from the hot springs where birds drank, or other cyanobacterial phytoplankton. Lindsay showed that the most likely causes – as shown in the 1970s – were infectious diseases such as avian tuberculosis and pneumonia.
Flamingos fascinate my fellow researchers
One of my co-workers was Dr Brooks Childress, a businessman who had retired to study his hobby: birds. After completing a PhD with me, he led the flamingo study. In 2003, Brooks found a dead bird on the lakeshore that had been ringed in 1962 at Lake Magadi as a chick, providing the first evidence of flamingos’ longevity. Later, searching the British Trust for Ornithology’s ringing recovery archive, he discovered a record of an individual found near Laayoune, Western Sahara, near some small lakes in sand dunes. That remains the only evidence of exchange between the main lesser flamingo population in East Africa and West Africa’s smaller population. This finding furthered our interest in flamingo movements and we secured funding to fit satellite transmitters to track eight birds captured at Bogoria and identify their feeding lakes. The data revealed locations in the three main East African countries and the time birds spent there (Figure 3).
Figure 3. The lakes and wetland visited by satellite-tagged flamingos.
We also used DNA fingerprinting of blood samples collected whilst fitting transmitters to compare the populations. Linking with new collaborators Graham McCulloch (Trinity College Dublin) and Guiseppe Crosa (Insubria, Italy), we compared the DNA of the populations in southern and East Africa. A dramatic upsurge in numbers in southern Africa coincident with decrease in numbers at Kenyan lakes in the 1980s had suggested movement between them, but no bird had been recorded from the countries in between. We confirmed the populations’ genetic similarity, evidencing frequent exchange. Later, we partnered with Dr Bhavbhuti Parasharya (Anand University, India) to analyse our blood samples and feathers he collected from the flamingo breeding site in India’s Rann of Kutch. The results indicated that these two populations had diverged ≥150 kya, but that similarities were maintained by a genetic exchange equivalent to 2–3 individuals per generation. This fits with ornithological observations of the species as an itinerant in all coastal states between India and Kenya.
Lesser flamingos only breed at one site in East Africa: Lake Natron in Tanzania. Elsewhere, they suffer predation, but Lake Natron is so large (80 × 32 km), hot, dry and dusty that predators cannot reach the lake centre. Here, flamingos build nest mounds of soda mud on ‘trona’: tiny islands of crystallised salt, but how and when they breed is little known. Whilst on a recce, a fantastic sight of ~750,000 birds feeding on the epipelic mud layer in the shallow (<6 cm) water of a lagoon inspired a revelation: Lake Natron was important for feeding as well as breeding.
Flamingos fascinate students too
I had started taking interdisciplinary science Masters students on a field course at Bogoria. One such student, Emma Tebbs, went on to complete a PhD in which we aimed to identify the full range of flamingo feeding lakes by their chlorophyll content, using satellite imagery. Emma developed an innovative algorithm to quantify planktonic and epipelic chlorophyll concentrations. Comparing her satellite-based chlorophyll calculations with field measurements made over two years by Steve Odour (Egerton University, Kenya) enabled Emma to produce a 13-year timescale for Lake Bogoria, the other two major Kenyan soda lakes (Nakuru and Elmenteita), Lake Natron and a large, remote pan in northern Kenya, Lake Logipi (Figure 4).
Figure 4. Lakes Bogoria (left) and Natron (right) showing true colour images with calculated biomasses.
Lesser flamingos had been thought to feed either on planktonic spirulina or epipelic biofilms, the latter being a poor alternative in very shallow water. Emma’s work had shown that at certain times, the areas providing this epipelic food in remote shallow pans can be well in excess of lake areas (Tebbs et al. 2015). Two students investigated this further.
One used Perspex enclosures to measure oxygen production and thus estimate the productivity of epipelic biofilms, and showed that, in nutrient-rich spring-fed lagoons such as Natron, it can be very high – comparable with phytoplankton productivity. The other investigated feeding behaviour in relation to planktonic and epipelic food availability, and demonstrated lake turbulence as an important influence on spirulina distribution. Lesser flamingos often cluster around the shorelines, but stand in water too deep for epipelic biomass.
This research found that spirulina often settles in sediments in shallow water, probably damaged by light exposure following calm days when it floats as scum, thus providing a rich feeding layer ~15 cm below the water surface.
Emma also used satellite imagery to interpret the conditions under which lesser flamingo breeding would be successful in Lake Natron. She compared changes in the lake area between 2000 and 2013 with times when breeding occurred, based on direct observations or the subsequent appearance of juvenile birds. This research showed that breeding was only successful when lake levels were intermediate and declining, whereas higher levels prevented trona islands from forming, and lower levels allowed predators to wade across the salt flats from dry land (Figure 5).
Figure 5. A true colour image of Lake Natron, February 2004, showing trona islands visible at intermediate water levels, suitable for flamingo breeding.
The future of flamingos
Much of what was then known about soda lakes was brought together by Schagerl (2015). But as we were preparing our chapters, enormous lake level rises happened in East Africa, destroying research infrastructure and thus severely compromising cross-disciplinary field research. Nevertheless, interpretation of satellite imagery is still possible, as shown by PhD researcher Aidan Byrne (supervised by Dr Emma Tebbs, King’s College London). Aidan has demonstrated that most soda lakes used by lesser flamingos have declined in their food biomass since ~2015, associated with these dramatic, hydrological changes (Byrne et al. 2024).
Lesser flamingos depend upon large networks of saline wetlands to find enough food, and they breed only in the few most extreme and isolated of these wetlands. Many are still pristine, albeit now affected by climate-driven hydrological changes. As a result, the lesser flamingo is near-threatened (sensu the IUCN Red List) and its populations are decreasing, making the protection of key breeding and feeding sites such as Lake Natron (a Ramsar wetland) a priority for its conservation.
Acknowledgements
My research was largely funded by the Earthwatch Institute and The Darwin Initiative, and also by many individual students and the British Council. My research permissions from the Kenya Government came through partnership with the University of Nairobi, National Museums of Kenya, Kenya Wildlife Services, Baringo County Council and the communities around Lake Bogoria. My research camps and logistics were managed by many Kenyans over 33 years, some of whom, as are three scientific colleagues, are no longer with us and are remembered in my writing.
References
Byrne, A., Tebbs, E.J. … Harper, D. et al. 2024. Productivity declines threaten East African soda lakes and the iconic Lesser Flamingo. Current Biology 34: 1786–1793. https://doi.org/10.1016/j.cub.2024.03.006
Harper, D.M. et al., 2003. Aquatic biodiversity and saline lakes: Lake Bogoria, National Reserve, Kenya. In: Martens, K. (ed.). Aquatic Biodiversity. Developments in Hydrobiology 171: 259–276. https://doi.org/10.1007/978-94-007-1084-9_19
Oaks, J.L. … Harper, D.M. et al. 2006. Septic arthritis and disseminated infections caused by Mycobacterium avium in Lesser Flamingos, Lake Bogoria, Kenya. Flamingo 14: 30–32.
Schagerl, M. (ed.) 2015. Soda Lakes of East Africa. Springer. https://link.springer.com/book/10.1007/978-3-319-28622-8
Tebbs, E.J. … Harper, D.M. 2013. Remote sensing the hydrological variability of Tanzania’s Lake Natron, a vital Lesser Flamingo breeding site under threat. Ecohydrology & Hydrobiology 13; 148–158. https://doi.org/10.1016/j.ecohyd.2013.02.002
Tebbs, E.J. … Harper, D.M. 2015. Regional assessment of lake ecological states using Landsat: A classification scheme for alkaline–saline, flamingo lakes in the East African Rift Valley. International Journal of Applied Earth Observation and Geoinformation 40: 100–108. https://doi.org/10.1016/j.jag.2015.03.010
Further FBA reading
The Freshwater Biological Association publishes a wide range of books and offers a number of courses throughout the year. Check out our shop here.
Get involved
Our scientific research builds a community of action, bringing people and organisations together to deliver the urgent action needed to protect freshwaters. Join us in protecting freshwater environments now and for the future.
