The ridge of mountains that runs parallel to Madagascar’s east coast descend in short order into rolling hills, followed by sand dunes, and ultimately meet the Indian Ocean. These rolling hills and sand dunes were once cloaked in low-elevation humid forest and littoral forest. Today those habitats are some of the most heavily affected by human activities, and only scattered tiny fragments remain. These fragments are strongholds for species that were presumably formerly widespread in this area, making them of great interest for research purposes, and extremely high importance for conservation efforts. However, the fauna (and flora) of many such fragments remain poorly characterised, or even entirely unstudied.
In November–December 2024, I took a team to conduct baseline herpetological surveys in two such areas in the Atsimo Atsinanana region of southeastern Madagascar. Our target was to perform herpetological surveys in the New Protected Area of Agnalazaha, which is managed by Missouri Botanical Gardens, and in Manombo Special Reserve, managed by Madagascar National Parks. The former has never been formally surveyed, and the latter has been visited only briefly and superficially perviously.
The team consisted of myself, my postdoc Alice Petzold (Uni Potsdam), Master’s student Kajsa Lundkvist (Uni Copenhagen), Bachelor’s student Christian Thrane (Uni Copenhagen), graduate researcher Sandratra Rakotomanga (Uni Antananarivo), graduate researcher Tanjoniana Rabarijaonina (Uni Antananarivo), and animal finder extraordinaire Angeluc Razafimanantsoa. We were accompanied by our driver, Nary, for most of the trip, and locally hired a cook, Louisette (who, as usual, wound up being the most important team member) as well as a host of local guides. We were also visited by two representatives of the local ministry of environment and durable development, who conducted audits of our research. At times we were some thirteen people, not to mention the battalion of porters who took us into and out of the forest in Agnalazaha. It takes a village.
Although our trip had a total duration of five weeks, in total we spent less than three weeks at our target locations. Part of this is because it takes two days of 14–16 hours in the car to cover the 700 km of less-than-ideal roads between Antananarivo and the Atsimo Atsinanana region. But we also took our time on our way down, stopping in Ranomafana for a few days to see some montane rainforest, practice some basic field techniques, and get the first bout of food poisoning out of the way. Whilst there, we had the great pleasure of bumping into Pat Wright, who invited us to the premiere of ‘Ivohiboro: The Lost Forest’ at Centre Valbio; and of working with Emile Rajeriarison, who gave us a tour of Valbio’s amazing entomological collections, as well as acting as our guide for some forays into the forest. This was especially valuable, as I was able to photograph a few frogs in Ranomafana that I have not seen previously.
So it was not until the 11th of November that we made it to Farafangana, the capital of the Atsimo Atsinanana Region. There, we made final purchases and announced ourselves and our plans to the local Ministry of Environment and Durable Development, and then made our way down to the MBG Office at Agnalazaha. After a short night in the bungalows, we finally made our way on the 14th of November into the forest.
Within just a few hours of hiking, we had arrived at our camp site; a cleared patch within the Strict Protection Zone of the protected area. Immediately we noticed that water was going to be a substantial issue; the only water nearby was standing swamp water, and even in the swamp the water level was very low. The rains were late, and there was no other water source here. If you have even turned to page one of most survival books, you will know that drinking standing water is a big no-no. However, treated appropriately and carefully, it can be drunk. And so we set up a chain, first filtering the water, then boiling the crap out of it, and then treating it with pur-eau, a bleach-based water treatment. What came out was an unpleasantly evocative pale yellow colour, but it was drinkable. However, over the next few days, the water level dropped still further, and at some point we had no choice but to organise for part of the team to go and purchase all of the water bottles they could find on the market, and bring organise a team of porters to bring in all of our drinking water for the next days; some seventy litres.
Agnalazaha protects a sizeable fragment of littoral forest. This forest is unlike anything I have ever seen. The canopy is low, just a few metres high, making the forest rather light during the day. Most of the forest is growing on sand, beneath a rather thin layer of leaf litter. Along the paths through the forest we found a huge diversity of orchids, a handful of euphorbs, and Madagascar’s only native cactus, Rhipsalis baccifera, and the ground was littered with clumps of dry moss growing directly on the sand. Some aspects reminded me of the spiny forests of the island’s south. But in between the forest-clad dunes (for lack of a better word, for they roll just like dunes) there lie swamps, where there is a thick layer of mud, and in the rainy season surely rather a lot of water. Here there are a variety of Pandanus species, as well as a lot of Ravenala (both groups important breeding grounds for frogs!). Amongst these swamps we were assured that there are often areas of quicksand, but I was disappointed not to encounter any of these—I wanted an opportunity to make use of all that preparation from my childhood. All of this makes for a remarkable diversity of habitats for reptiles and amphibians, and consequently an eclectic assemblage of herpetofaunal species.
Our surveys in Agnalazaha lasted until the 21st of November, and covered a considerable area of ground. Angeluc in particular went on numerous forays during the day, always bringing to camp a host of goodies from his adventures. In total, preliminary results indicate that we found 17 amphibian species and 21 reptile species in Agnalazaha, 47% and 30% of which, respectively, were new records for the protected area. The paucity of amphibian diversity owes to the lack of rain—by the end of our visit, our camp was a dustbowl—as well as the general absence of any large, flowing rivers with bordering riparian forest within the protected area; all of the frogs we encountered breed either away from water (nesting in phytotelms, jelly nests, or otherwise) or in standing water (e.g. Boophis subgenus Sahona). The lack of substantial bodies of flowing water makes it impossible for most Boophis, and many other mantellids, to occur.
We had planned to stay in Agnalazaha until the 22nd, but the threat of heavy rain and a rapidly rising water level led to the decision to leave the forest a day early. In the end, this would not have been necessary; although it did rain a little, thankfully, it was not enough to be of any concern. But regardless, we left the forest on the 21st, and moved to the Special Reserve the following morning.
In Manombo, the park entrance there is a set of concrete buildings, with shelters for tents to be built under, and a relatively large field behind it. The buildings double as roosts for what must an absolutely massive number of bats, which we could hear—and smell—at all times. Unfortunately the tent shelters were all occupied by several groups of other visitors, who were apparently conducting surveys in nearby villages. So we had no choice but to set up our tents elsewhere. It took a few attempts to choose a good location, but in the end we settled for putting the tents close together in the middle of the field. This unfortunately meant that they were totally unshaded over the course of the day, even though we did put up a tarp to protect them a least a little in the afternoon. When it was 36°C outside the tents, the temperatures inside them were deadly to humans and electronics alike; we measured an air temperature over 42°C in the Food/Equipment tent during the day. So we were left with no choice but to bring out all our electronics and keep them in the shade during the day. Fortunately, the position did give us access to a protected shelter, where we were able to have our field lab and eating space, with shade and a solid floor; it was a tolerable compromise.
Once we had settled, there ensued a number of changes of plan; these were too numerous and tedious to enumerate, but as a consequence it was some time before we were able to operate at proper capacity in the forest, and especially to go into the forest during the day; we had arrived just before a weekend, and this complicated matters. We finally settled on focussing all of our efforts on the humid forest patch of Manombo, which, to my knowledge, has never been surveyed in earnest before. For the first days we worked in the forest only at night, because work during the day on the weekend was not possible. We also had a 23h curfew imposed on us, which meant every trip into the forest was very brief indeed.
The reason for these restrictions is the threat of dahalo encounters. The dahalo, also called malaso, are a semi-organised group of bandits, based originally in cattle rustling, spread across Madagascar. They are concentrated especially in specific areas, and Atsimo-Atsinanana seems to have a higher concentration, or more conflict, than most areas. Such areas are referred to as ‘zones rouges’. In the case of Manombo Special Reserve, the dahalo (who are local villagers) are apparently involved in the harvest of wood from within the reserve, which they do mostly at night. Consequently, it can be difficult to work in the forest at night; we had to take two local guides and one forest guard with us every time we went into the park, and had this curfew to contend with. In the end, I was satisfied with this compromise; I had originally been worried that we would not be able to reside anywhere near the forest, and wouldn’t be able to go in at night at all.
Immediately as we walked into this forest, it was obvious that it is quite significantly different than Agnalazaha. Firstly, it is steeper, and the soil inside the forest is only sand in patches, but mostly soil and mud elsewhere. There are large dry riverbeds with massive exposed rocks. The humidity was higher, and the temperature substantially lower. There are stands of Pandanus of a species I have never seen before that grows at least seven metres in height and makes a canopy of screwpalms above, with a deep layer of razor-sharp fronds below; a bizarre habitat absolutely brimming with amphibians that were physically dangerous to try to catch.
Every night, and eventually during the day as well, we visited a different site within this forest (only revisiting a few for specific targets). Sites of entry were spread over the eastern edge of the protected area, and hit quite a few different habitat types. On one day, we undertook a more major trip to visit the natural pool, or piscine naturelle, located much deeper inside the park. This was a several hour hike, but it was hugely rewarding. In relatively short order, we added eight or more species to the list of taxa we had found on the trip, because here there was a lot of water and a massive river that, though it was not really flowing, was fresh and clear. Still we did not find the Boophis that occur around faster flowing water, but this was more than made up for by some of the other finds, which included the incredible Paradoxophyla palmata and the bizarre Gephyromantis (Laurentomantis) malagasius and a beautiful, still to be identified Lycodryas species.
We were fortunate that in the last couple of days at Manombo, the weather changed, and we got some significant rain. This immediately brought out some frogs that we would never otherwise have encountered. Most significantly among these were the Anodonthyla that were the target of Kajsa’s work; and Plethodontohyla alluaudi, a charmingly round microhylid that only calls during and immediately after rain—where it also reveals itself to occur at surprising densities.
It was not until our last night that we were brought to an area with faster flowing water, where we suddenly encountered Boophis (Boophis) viridis in massive numbers, emitting the loud pops and whistles of their advertisement calls. I was disappointed not to find some of the other stream Boophis which had previously been recorded from Manombo, but later was able to explain this—those species had been found much further down this same river, where it gets much larger, and where there are tall trees on either bank.
In total, preliminary results indicate that we found 38 amphibian species and 25 reptile species in Manombo Special Reserve, 41% and 44% of which, respectively, may be new records for the reserve. Of these, several may represent new species; I am excited to be able to get this material to Copenhagen and get stuck into the genetic sequencing of the material.
On the 2nd of December we packed up camp and left the field to make it to an appointment we had made in advance with the local ministry, with the hope and goal of completing the necessary steps to transport the material we had collected to Tana the following morning. This was not to be, however; paperwork challenges pertaining to interpretation of terminology had to be clarified, as well as some other matters. To keep a long story short, it was not for two days that we were finally able to depart from Farafangana and make our way first to Ranomafana, and then onwards again to Antananarivo. This left us with precious little time in Tana to do the last things we wanted to do. But somehow we managed to squeeze in most of the important things, like sorting the specimens. We had originally wanted to devote some substantial time to working in the collections of the Mention Zoologie et Biodiversité Animale, but in the end we could only devote a few short hours to working there—which were nonetheless very valuable.
The trip was a whirlwind, but in the end, all ran relatively smoothly. We had some really great successes, and these showcase also how valuable such baseline surveys are, and how little we actually know about the diversity and distribution of herpetofauna in Madagascar. It will certainly be necessary for others to conduct further surveys in these areas, especially later in the rainy season—but at least in Manombo, this is going to remain limited while the dahalo issue persists. So our knowledge of herpetofauna in the precious low-elevation forest fragments will remain, well, fragmentary, for the foreseeable future.