Studying the birds of Anjouan

ECDD ecological technician Amelaïd Houmadi is studying for a distance learning Masters from the Université Libre de Burkina Faso and has just finished the fieldwork for his research project into the birds endemic to the Comoros on Anjouan. To celebrate he’s put a few pictures together for us of some of the striking birds he’s been seeing (and hearing) in the forests of Anjouan…

We are very grateful to the anonymous donor to Durrell Wildlife Conservation Trust who agreed to fund the cost of this project.

Livingstone’s fruit bat survey video diary 4: The last Anjouan roosts

The team visit the last of the 16 Livingstone’s fruit bat roosts in Anjouan, finding more evidence of the threats that extreme weather poses to the survival of the species now that so little of its habitat remains. The last surveys take them to some of the most difficult terrain on Anjouan at Lingoni and Nindri, and a welcome relief with the easiest walk to a roost at Adda.

Dindri and Mpage: a troubled roost finally abandoned, and a healthy site faces a new threat

Livingstone fruit bat Comoros

Amelaïd and local school teacher Toimi who led us to the roost site at Dindri

Livingstone's fruit bat Comoros

A tree stump near the Dindri roost

This week Amelaid and I headed to the village of Dindri on the east of Anjouan to visit the nearby Livingstone’s fruit bat roost. Accompanying us was a local school teacher Toimi who takes a keen interest in the bats. The walk to this roost is one of the least demanding. The land is so heavily cultivated that we don’t even walk through any forest before arriving.

We didn’t set off with high expectations; previous counts have shown a big decline in numbers of Livingstones at this roost. Toimi explained that the area used to be covered in big trees but since the millennium there’s been lots of felling for firewood and construction, including the principal roost tree for the Livingstone’s fruit bats. There was also a big landslide here soon after large trees were felled. Toimi told us “In the past people didn’t cultivate this far from the village. Now the landscape has changed and villagers also have to walk a long way to find firewood.”

However, on arrival at the roost, even our low expectations were dashed when didn’t find a single Livingstone at the roost. It seems they’ve finally abandoned this roost altogether and left it over to the much more numerous (and noisy) fruit bat Pteropus seychellensis. A sign of how degraded the forest has become was that there were even tomatoes planted nearby, which are not shade crops like the banana and taro that we’re used to finding planted under forest trees.

Livingstone fruit bat survey Comoros

Livingstone’s fruit bat’s cousin Pteropus seychellensis comorensis is found on all four islands of the Comoros archipelago, as well as Mafia Island off Tanzania. It is much more abundant and widespread than the Livingstones, including throughout towns and villages.

After the disappointment of the abandoned Dindri roost, the next day we launched straight in to one of the longest climbs up to the roost at Mpage. Amelaïd and I thought we knew where we were going, but after climbing up the ridge for over two hours we were met by a very different landscape to what we remembered from the last time we were here six months ago. A lot of trees had been felled, making it difficult to find the right paths. I lost some of my love for the forest whilst scrambling around the slopes searching for this roost for nearly an hour with branches whipping across my face and getting stuck in my rucksack pulling me back.

Livingstone fruit bat survey Comoros

The field that has just been cleared right next to the Livingstone’s fruit bat roost at Mpage, Anjouan.

Once a decent viewpoint was found, we counted 36 Livingstone’s fruit bats at the Mpage roost.

We finally caught sight of the bats when we found ourselves in a field above the roost trees which had been cleared so recently that the tree stumps were still weeping sap. From our last visit I remembered this site to have been a text book roost site in pristine forest, so it was a real shock to see the change. It’s so far from the village that the farmer has built a hut next to his field, so he doesn’t have to climb up every day.

My immediate reaction is anger at what is happening to the forest and horror at the thought that the bats’ roost tree is next in line. But when you meet the individuals responsible you simply see someone trying to get by and take enough home to feed his family.

This is why the projects work supporting sustainable methods of intensifying agricultural production is so important. With the right skills, tools and knowledge Comorian farmers should be able to produce enough crops in existing agricultural land without needing to climb even further into the forest in search of fertile soils.

After carrying out our count of the bats at the roost, we managed to speak to the field owner and were relieved to hear that he does not plan to fell the big trees that the bats use because they hold the soil together for his crops. He also stated that he knows that they disperse seeds and that it is thanks to the bats that he has a forest to work in.

This is encouraging because the Mpage roost appears to be a bit of a bat nursery. At first sight the individuals looked unusually big, but when they opened their wings to warm up in the sun’s rays we realised that the majority had infants clinging to their chests. In total we counted 36 bats, and of those, 22 were carrying infants.

The rapid clearance of forest for cultivation shows how crucial land management planning in Anjouan is, so that food can be produced in certain areas and important areas of forest protected. No official protection exists either for this species or for the habitat it depends on, and the forest is disappearing fast.

Livingstone fruit bat baby

When the bats at Mpage stretched their wings in the sun we noticed a large number were carrying babies clinging to their chests.

Video diary 2: Limbi and M’romaji roosts

Bronny and Ishaka’s adventures visiting the Livingstone fruit bat roosts at Limbi and M’romaji: camping in a classroom to escape excited children, balancing over a precipice to get a view of the bats and a noisy roost full of the Livingstone’s fruit bat’s cousins Pteropus seychellensis. Oh and a bat falling out of a tree… Don’t worry there’s a happy ending.

The only mzungu in the village

Livingstone's fruit bat camping

Our lodgings for the night

This week it felt like we’d got into the rhythm of the surveys a bit more… maybe my legs had toughened up a bit, or maybe it was because we found some better news than in the disappointing first week. Ishaka and I were heading to visit two roosts I had never been to before, first to one near the village of Limbi, and then to M’romaji, so I was excited, not knowing quite what we might find.

I think the trip to Limbi will hold some of my fondest memories of this study. It seems our arrival in the village was a major event for the community. We were told that this is only the second time a mzungu (foreigner) has stayed in the village. After explaining who we were and what we were doing, the villagers insisted that we pitch our tent in a school classroom for the night because the children were too excited and wouldn’t leave us alone if we were outside. They were still pretty active though, clambering around the windows to get a good look at the weird white lady pitching a tent in their classroom.

Livingstone's fruit bat Comoros survey

Athoumane, the owner of the land perched on the dead tree at our vantage point

Livingstone's fruit bat Comoros survey

Ishaka and Bastoine counting the bats from our precipice

When we turn up to survey a roost we always try to make sure we are accompanied by people from the village who’s land the roost is found in, so that they can see what we’re up to and we can find out their perspective of how the population is doing and the threats to the roost. But it’s not always easy to find people willing to come up with us. This wasn’t a problem in Limbi though and we had two willing volunteers: Bastoine, who used to survey the roost for a local NGO and Athoumane, who owns the land.

This is one of the tougher roosts to get to. It’s on a particularly steep slope (even for Anjouan) in an area of dense natural forest – good news for the bats, but less so for researchers trying to find a vantage point. Our hearts dropped when we arrived at our destination and found a big cloud was blocking our view of the bats… The thought of having to climb up here again didn’t bear thinking about.
Luckily the cloud soon dispersed and we could see some bats hanging peacefully in beautiful forest – a very welcome sight. In the end we found the best view balanced on a fallen tree over a precipice, keeping our eyes on the bats and not looking down!

The topography of the site suggests it’s not at a high risk of deforestation. But despite this Athoumane has still managed to plant banana plants under the roost. This field is important to him because his other fields are not fertile enough to grow bananas and taro. Fortunately for the bats though he is happy that they are in his field and wants to protect them so that they don’t go elsewhere.

After counting 46 bats, we half walked half slid down the mountain in high spirits, finally having found a healthy roost.

Livingstone's fruit bat Comoros survey

Livingstone’s fruit bats stretching their wings to sun themselves at the Limbi roost

From Limbi we went on to a roost not too far away at M’romaji. This roost is completely different to the other sites around the island as it’s at the base of a ravine rather than on a steep slope. Ishaka says the roost used to be even closer to the village but moved a few years ago due to disturbance and deforestation.

It’s a noisy roost, but not because of humans but because another species of bat, Pteropus seychellensis, which also inhabits this area in much greater numbers. The seychellensis are far more active and noisy then the larger, tranquil Livingstones. It seems the Livingstones don’t mind the noise here though and numbers have not changed since the last surveys.

Livingstone's fruit bats Comoros survey

Ishaka and Nahouda looking at the M’romaji roost

We were accompanied by Nahouda, who is the head of the village and owns the land around the roost. He shares four fields with his nine brothers, and relies on agriculture for his living. He is obviously passionate about the bats though, and told us that he is the “guardian of the bats” and stops his family from felling the roost trees. When we asked why he said “Because they’re important for Anjouan and the Comoros – they’re only found here! We should plant lots of trees to make sure they don’t disappear.”

After mine and Ishaka’s successful trip, we got back to Mutsamudu and found good news waiting for us from Amelaïd and Daniel’s venture to the Hombo roost. The bats at this roost could well be the cousins of those in the ‘safety net’ populations at Durrell Wildlife Park and Bristol Zoo, as it’s not far from where the first animals were captured. On their trip, Daniel and Amelaid counted almost double the number we observed in the wet season six months ago – a whopping 167 individuals.

Livingstone's fruit bat Comoros survey

A kapok tree near the Hombo roost where bats were captured for the Durrell (JWPT) captive breeding programme

Fruit bats lost and found: mixed news from the first roosts of the survey

The period in the run up to the start of these surveys proved that even the most thorough plans are made to be adjusted and readjusted… first it was me breaking my toe, and then it was torrential rains just as the ‘dry’ season was supposed to be starting. But finally we were ready, and could get started on the surveys. As we were already behind schedule, there was no time for a slow warm up so we launched straight in, covering four roosts in the first week.

Livingsone fruit bat research team

Daniel, Ishaka and Amelaïd tucking in to bananas and fish

We set out all together as a team of four: Ishaka, Daniel, Amelaid and I, so that we could refresh the methodology together, making sure that we’re consistent in the way we count the bats at each roost. For the next outings we’ll split into two groups so we can cover all fifteen roost sites in Anjouan in as short a time as possible, to make sure the results are comparable (it’s not a good idea to compare a count from June with a count from September for example, as the populations at each roost fluctuate throughout the year).

Before starting these surveys we studied roosts over a whole day to work out the best time to carry out our counts – we want to count the bats at the time when the roost is busiest. We found this was between 8am and 11am when the bats are resting before they set out to forage in the afternoon. So normally we slog up to the roost in the late afternoon, and set up camp not too far away so that we are in place and ready to start the count in the morning.
It’s been a while since we were all in the field together so sprits were high as we set off up the mountain to the first roost in Salamani. But after the sixth person passed us heading down from the forest with huge planks of wood on their heads we knew this wasn’t a good sign for the bats.

When we got to the roost we found it had changed a lot since our last visit in December: more crops had been planted directly under the roost, and a whole new banana field had been cleared nearby. It really shows how quickly the landscape is changing. Ishaka has been counting the Livingstone’s fruit bats for over 15 years and remembers this site to have been dense forest only ten years ago. Since then, as well as agriculture moving in, some of the original roost trees have been felled. The bats have moved slightly away from the intensifying agriculture, but luckily they haven’t abandoned the main roost tree.

Forest clearance Anjouan Comoros

The newly cleared banana field by the Salamani roost

After checking out the roost we settled back down to the all too familiar field rhythm of setting up camp and eating our dinner of cold fried fish and bananas, before bedding down in our tents to be ready for the morning counts.

In the morning we counted only 40 bats, which is worrying as we counted 97 during the wet season; the bats don’t seem to be coping well with the disturbance around the roost. With the first count behind us I was slightly relieved: I seemed to remember the climbing being harder last time… However part way back down and the familiar shaking knees returned as I slid down the slope. It’s the downhill that always gets me.

The next day Ishaka and I set off for the roost near the village of Ouzini, a village high in the mountains, an hour’s walk from the nearest road, while Daniel and Amelaid headed off to count the roost near another mountain village called Outsa.

Landslide Ouzini

This landslide tore through a farmer’s field in Ouzini

Livingstone fruit bat roost hit by landslide Anjouan Comoros

The landslide that swept right by the Livingstone’s roost at Ouzini but luckily didn’t hit any of their favourite trees

Back in May the island suffered from some terrible storms, with flooding wiping out many houses, roads and water supplies. The equivalent of a whole year’s rain fell in four days. The area around Ouzini is particularly steep, even for Anjouan and suffered many landslides following the flash floods. It was quite something to see the boulders that had been carried down the slopes, obliterating entire fields and some farmers’ entire crops. By the time we got there though we found villagers hadn’t wasted any time in planting banana plants in the newly turned soils.

Ishaka and I were not ready for what awaited us at the Ouzini roost. One of the landslides had swept right through the centre of the cluster of trees the bats roost in. This site is so steep that the trees are not likely to be targeted for construction materials because there’s no way of carrying the planks out, but they were hit by storms instead.

By some small miracle the landslide had missed the particular trees the bats choose for roosting and we found 72 bats in the same 14 trees that we did back in December, much lower than the 105 from last time. The field owner was busy chopping up the fallen trees for firewood, creating a lot of noise in an area which was previously undisturbed. We don’t know how the change in physical environment will affect the colony, but fingers crossed when the noise stops, it will be calm enough for the other bats to return again.

Mongoose lemurs Anjouan

Inquisitive mongoose lemurs came to see what we were up to as we were searching for the new roost

So it was depressing news from our first two sites, but every cloud has a silver lining and back in the village of Salamani we heard rumours of another roost that has previously never been counted, found far, far into the forest. So we were all excited to explore further into the forest in the hope of discovering more bats.

We had been warned that the new roost was far away, but after an hour of scrabbling up an incredibly steep slope I started to wonder if it would be my heart or legs that would give out first. We finally found nine big black bats lazily dangling in a forest tree. Even though it’s a small number it’s a relief to find a new site, and this brings the total number of roosts in Anjouan up to 16.

Livingstone fruit bat roost Salamani Anjouan Comoros

The new roost we found deep in the forest above Salamani