Inbar
News Mapping Madagascar’s Bamboo Forests with INBAR and Tsinghua University

Mapping Madagascar’s Bamboo Forests with INBAR and Tsinghua University

INBAR News
22Feb

We find out what is was like to be part of the first ever project using remote sensing to map the bamboo cover on the island of Madagascar.

INBAR’s flagship project the Global Assessment of Bamboo and Rattan, or GABAR, is dedicated to expanding and updating the current knowledge base on bamboo and rattan to allow the global community to make the most of these amazing plants. As part of the project, INBAR and Tsinghua University, one of China’s top universities, published the first every remote sensing assessment of Madagascar’s bamboo resources, leading to detailed maps showing the location of bamboo forests across the country.

Madagascar, an INBAR member state, has been previously shown to be the most diverse in Africa in terms of different bamboo species. Bamboo is a culturally and economically important crop in Madagascar, and current INBAR efforts focus on working with communities to sustainably exploit their bamboo resources and relive pressure on the precious rainforest, firewood from which currently provides an energy source for 68% of the population. Madagascar’s bamboo forests are also home to two critically endangered animals –  the greater bamboo lemur and the ploughshare tortoise (pictured below). Although Madagascar’s forests are often considered a global conservation priority, there was previously almost no accurate information available on the country’s bamboo forest distribution. So for INBAR, the Tsinghua university team and their local partners, the need for this study was there: it was up to them to make it happen.

The rare ploughshare tortoise is only found in the dense bamboo forests of Madagascar

Using remote sensing from satellite imaging to map bamboo resources was nothing new for the Tsinghua team, originally made up of Yuanyuan Zhao and Duole Feng. Working with INBAR staff members Durai Jayaraman and Daniel Belay, the team mapped the bamboo forests of Ethiopia, Kenya and Uganda back in 2016. For this study, Yuanyuan Zhao was joined by Meinan Zhang, who had been interested in bamboo mapping since her first year at Tsinghua university. Now a researcher in data science, specializing in developing algorithms for land cover mapping or natural resources, she worked with a similar project in China’s tropical Hainan island before joining the team that travelled to Madagascar for the fieldwork portion of the study. She shared with us her experiences of the trip.

“I traveled with a postdoc from our research group, who was mainly responsible for translating – I can’t speak Malagasy or French, the main local languages! Aside from myself and the translator, we had three co-workers from the Missouri Botanical Garden in Madagascar. They were local and really friendly, and provided lots of help during the field. It wasn’t just work – we also built a good friendship with them!”

The field team in Madagascar

The team traveled throughout eastern and northern Madagascar to gain a closer understanding of the land cover and climate of the country, collecting 1318 bamboo samples and several other land cover samples. The hands-on, intensive nature of this work was a chance for the team to learn more about Madagascar, one of the most distinctly biodiverse countries in the world, and to appreciate the significance of bamboo for Madagascar’s rural communities.

“The experience was brilliant! I was really amazed by just how much biodiversity I saw! For the first time, I got a real feel for how important bamboo is for people living in the forest. We met farmers who make their living from bamboo, who told us through the translators just how critical it was for their family’s  survival. Seeing the hope in their eyes brought to life how important our project was.”

Meinan Zhang from Tsinghua University collecting field samples

Carrying out this type of study with bamboo carries its own particular set of limitations. When undergoing the remote sensing satellite imagery work, pixel resolution for the imagery was set at 30m x 30m. So areas of bamboo that were smaller than this pixel size may not have been picked up by the satelite images that were used by the Tsinghua team to assess bamboo forest cover. This limitation of the study was an important realization for the team, as most of the bamboo grown in smallholder farms in Madagascar appeared to be in plots smaller than 30m x 30m.

In addition, measuring bamboo cover as separate from other forest cover isn’t easy: with so many different species of bamboo and different types of land cover, differentiating what was and wasn’t bamboo was a challenge due to the similar spectral properties of the remotely sensed images f different plants or forest. But the team developed an appropriate research methodology for Madagascar to obtain detailed maps of bamboo forest cover by focusing on the two predominant bamboo species, Bambusa Vulgaris and Valiha Diffusa.

Overall, the team found 11236.94km2 of bamboo forest cover in Madagascar and created high-resolution maps like this:

 

Scientific research projects like this are a major part of GABAR and make important contributions to the global information database which will have far reaching impacts for biodiversity conservation, livelihood development and ecosystem services management. But for Meinan Zhang and her colleagues, it was also a chance to connect with another country and culture, share expertise and knowledge – and to find out why it is so important to study how and where bamboo grows.

To read the final report that INBAR and the team wrote and find out more about the study, read here.

For more information about GABAR, please contact coordinator Trinh Thanglong at ttlong@inbar.int