Wildfowl & Wetlands Trust (WWT)

Today was our final day in the field, and it was definitely a scoring day for the mergansers. Following the heavy rain last night, for the second consecutive evening, the Sao Francisco was murky with sediment that had been washed into the river. We wondered whether this might mean the mergansers would be hard to find, as when the river conditions are like this it is likely that they find foraging difficult and may therefore move up tributaries where the turbidity is likely to be lower. Sure enough, we only had a brief sighting of a single bird flying upstream, giving no opportunities for capture during the entire day.

When Leandro and I were paddling past one of the more significant tributaries in this section it was clear that this stream had not been affected by the rain in the same way as the Sao Francisco. Its water was almost as clear as it had been before the rain. Renata and Augusto searched this stream, but found only two muscovy ducks and a bevy of kingfishers.

So, despite the considerable successes of the past week, the mergansers had the last laugh, reminding us again that we still have so much to learn about how these birds use their habitat. This is hopefully what the Terra Brasilis team will now be able to study in more detail, by following the radio-tagged birds over the coming months. Further updates on these birds will be provided on the WWT website as information becomes available.

However, overall the past week has been extremely successful - we have caught and ringed the first 14 Brazilian mergansers to be marked, fitting five of those with radio transmitters which will now continue to provide valuable information on their habitat use, and the way the territories are arranged on the Sao Francisco, which to my inexperienced eyes seems surprisingly complex. The trip has also been a valuable learning experience for the team from Terra Brasilis, none of whom had any experience of catching mergansers before this trip, apart from the three birds that were mist-netted back in March.

They are now ready to continue to capture and mark more mergansers as the breeding season comes to an end for another year over the next month or so. This will include the pair on the River Peixe, close to Sao Roque, which we caught in March with one of their offspring from 2007. Their six young from this year are still only about half grown, and thus too small to ring, but should be ready within the next two weeks.

Sadly, I will be back in Britain by then, dreaming of Canastra, but the Brazilian merganser research and conservation effort here will thankfully continue in the able hands of Terra Brasilis. Partnerships such as these are one of the ways WWT can most effectively support the conservation of threatened waterbirds, and the efforts being made here to secure a bright future for one of the rarest birds in the world will continue to be supported by WWT in the years ahead.

With the weekend behind us, we had renewed hope of locating some mergansers. The priority was to try to locate the pair with eight young in the area close to Casca D'Anta. We split up to cover the same river sections as before; I walked the Cachoeirinha, alone this time as Ivana left us yesterday to return home. However, we were also joined by Adriano Gambarini, a professional wildlife photographer who has supported the Brazilian merganser project in the past by taking many fabulous photographs of the mergansers.

The weather had turned hot again, following the cooler and cloudier weekend, and it was a great start to the day to see a White-eared Puffbird sat on wires by the roadside. I set off down the Cachoeirinha, first checking upstream of my start point - an area where access is not permitted due to the presence of an active diamond mine. This is where the Terra Brasilis team believe this pair is likely to have been in recent days.

But not today - about 45 minutes later I was scanning a straight section of the river when two mergansers swam right out in front of me from behind a rock. Fantastic, but which birds were they? At first it seemed as if there were only two, which did not help to solve the mystery of the missing family, but then some young birds appeared and it was clear that this was the missing family. However, there were only six young birds, not eight, so it appears that two young birds have perished in the past two weeks, since these birds were last seen. This is probably not usual, though currently we have very little understanding of the extent and causes of juvenile mortality.

After enjoying great views of the family fishing in the rapids, a plan was hatched to attempt to catch them. Augusto came to replace me on watch duty so that I could go and help put up the nets. Once this was done, Augusto gently moved the birds downstream until they reached the nets. In order to maximise our chances of catching the whole family, we had put two nets across the narrow stream, one behind the other - the idea being that if any got through the first net, we had another chance of catching them. Unfortunately, only one bird, an adult, was caught in the first net, and just two of the remaining seven, both young birds, were caught in the second net. The other five managed to evade capture in both nets!

So, whilst it was great to catch another three birds - only the sixth, seventh and eighth individuals to have ever been ringed - it was frustrating that so many had escaped. After the three birds that had been caught had been ringed, measured and released we examined the nets to try to see how the birds had escaped. The base of the net was still completely pinned to the river bed by rock's (so they could not have got under the net) and there were no holes in the net either. This could only mean one thing - that the birds were able to squeeze through the normal mesh size of these nets, which to date had only been used to catch full grown and flying adults.

So, it would seem that the smaller young, and the slighter (because they have short wings) moulting adults were able to go right through these nets. Whilst this was obviously very frustrating, at least we had not done anything incorrect when setting the net - the only error was in using nets with a mesh size that was too big. However, as no one had ever caught Brazilian mergansers during the breeding season before this trip, we could perhaps be excused this indiscretion.

After a short lunch break at a roadside bar with a stunning view of the Canastra escarpment from its veranda, we decided the spend the afternoon trying to locate some other pairs known to be in the area. I accompanied Renata on a walk of the territory of the birds caught on the first day, whilst Ricardo (a local vet who has also been helping us) and Flavia boated down the area holding the pair with seven young. Renata and I had an extremely enjoyable afternoon, seeing some fantastic wildlife and landscapes, and also managed to relocate the pair of adults ringed last Friday.

We had great views of these birds as they foraged in the rapids before resting on some rocks, showing us their rings. However, we did not see the family, and it seems likely that they have also moved up a smaller tributary for the time being. Ricardo and Flavia had more luck, finding their family towards the end of their territory. However, what was a family of seven young two weeks ago is now a family of three young. So, like the family seen earlier on the Cachoeirinha, it appears that juvenile mortality in the latter stages of growth can be significant. Understanding the causes of this mortality, and whether it is related to human activities, is an important part of the further research that needs to be carried out here in the near future.

The day ended with a spectacular thunder storm, and great views of an Armadillo as it crossed the road in front of us on our way back to a very wet Sao Roque. Overall, a great day today with three more birds marked, and a number of observations of mergansers. The frustration of the nets still niggles, but we can easily resolve that problem for next year...

The weekend brings a lot of people to the Sao Francisco area, walking to Casca D'Anta, fishing and fishing and swimming in the river. Therefore, disturbance is greater and the mergansers are often harder to find.

We decided, therefore, to look for the same family as yesterday as this territory is in the upper reaches of the Sao Francisco and although this means it is closer to Casca D'Anta, it is further from where most people live and so there are fewer people fishing. The other pair, with seven young, that we also hope to locate and catch next week is closer to the nearby town of Vargem Bonita, and so we will look for these birds next week, or possibly tomorrow.

The day began in a clearly different mood to every other since I arrived - there was a thin layer of cloud over most of the sky, and a cooling breeze. When we started out, at 6am, I even needed a light fleece! We drove close to Casca D'Anta and I set out to walk along a tributary of the Sao Francisco, the Cachoeirinha, with Ivana. It was very different to the main Sao Francisco - at 5m about half the width and with a closed canopy of trees reaching from each side.

We saw some great birds, such as the hyperactive sharp-tailed steamcreeper, and some stunning butterflies too, but no mergansers. In fact, the only mergansers seen by the team today were again the two that we caught yesterday - Augusto observed them close to where they were captured, looking none the worse for their close encounter with pesky scientists on the previous day. So, still no sign of the family - where could they be?

Given that it was still only about 11am, and the only other birds to look for was the family with seven young which we were not going to look for on a Saturday, we decided to do a few other useful things such as test the boat that we will use for surveys of the lower reaches of the Sao Francisco, as it has not been used before and several people had not used such an inflatable to paddle down these rivers before. We then went for some lunch at a little rural restaurant with a fantastic view of the Canastra escarpment, and looked over maps of the whole area to see and discuss where surveys have been carried out to date, and identify the priorities for future surveys.

By then, it was time to head back to Sao Roque via a few stops to check lower reaches of the Sao Francisco, but as expected there were quite a few people fishing and swimming, and no mergansers to be seen.

We were up at 5.15am, it was light soon after - another cloudless and hot day loomed (and duly materialised) and we set off for the Sao Francisco. We stopped at a bridge at the downstream end of the territory of the birds we hoped to catch and Augusto set off walking along the riverbank to locate the birds.

He found them very close to the bridge, further downstream than where we had hoped to catch them. Therefore, he slowly moved them upstream until they were above the point where we wanted to set the net. We duly set the net - a large meshed mist net pegged to the bottom of the riverbed with rocks - and retreated to our positions, myself and some other people by the net, and others about 50m upstream to give a final bit of encouragement to the birds to go into the net.

Augusto began to move the birds towards the net and, after a short period when they were reluctant to proceed because of a group of black vultures that were disposing of a dead cow by the side of the river, they moved closer and closer (and I became increasingly nervous and excited!) until they passed the people 50m from us and made a final charge into the net!

All four went into the net - a real bonus as it later turned out that both adults could fly and could have easily flown over the net. However, their desire to remain with their young meant that they remained close to the river. Unfortunately, one juvenile managed to pass the net - probably by getting underneath it, but the other three were safely caught.

We ringed and measured the birds on the riverbank, and also fitted a radio transmitter to the juvenile. We had hoped to also fit a transmitter to an adult, but both had active moult in their tail, so this was not possible. Nevertheless, these were the first Brazilian mergansers to have been colour-ringed, marking another significant step forward for the conservation of this critically endangered species.

After the birds were released, they swam off downstream and were later seen with the second juvenile that had evaded capture. A great start to the trip - and it was still only 11am! We then moved to another territory of an unusual group - two adults and their seven young, plus another two adults, possibly young from the previous year that have remained with the adults whilst they rear another brood.

We looked at some places where we might try to catch these birds tomorrow, but we did not manage to see them - maybe tomorrow...

I arrived at an unusually hot (for the time of year) and sunny Belo Horizonte airport in the late afternoon, and was met by Livia Lins, from Terra Brasilis, the NGO with whom WWT are closely collaborating on Brazilian merganser conservation.

Despite an almost total lack of sleep in the past 36 hours, I was feeling wide-eyed at the new sights and sounds. We drove into the busy city centre, past avenues of spectacular Ipê trees, now covered with blooms of bright pink or yellow flowers. Whilst we were on the road the sun set like a stone, and it was fully dark by the time we arrived at the hotel where I was to stay tonight.

I managed to stay awake long enough to go into town and enjoy a superb and gargantuan barbeque, but was asleep within moments of returning to the hotel.