LESELAM: combating soil erosion in Mayotte

FIGHTING SOIL EROSION AND THE SILTATION OF THE LAGOON IN MAYOTTE MAYOTTE REGIONAL OFFICE

AN EXCEPTIONAL LAGOON

Mayotte's lagoon was listed a marine natural park in 2010. Its exceptional biodiversity makes it one of the planet's most remarkable lagoons. But this biodiversity is threatened due to siltation of sediments from Mayotte's natural, agricultural and urban areas. Mayotte has known big changes throughout its history, mainly agricultural, with major shifts in crop production particularly regarding ylang-ylang and sugar cane crops. But since the '90s, it has seen other kinds of big changes related to its exceptional demography, which has impacted naturally fragile environments. This fragility is based on the island's morphology, with its prevalent sloping surfaces, its tropical climate and intense rains, and its fragile and erodible soil.

THE LESELAM PROJECT

In 2012, the DEAL in Mayotte established a strategy for erosion to try to contain the issue of erosion and siltation in the lagoon. The LESELAM project to fight soil erosion and siltation in the lagoon comes within the scope of this plan, for its objective is to evaluate the inflow of sediments from Mayotte's drainage basin and to accompany individuals and professionals by implementing better conduct to contain this problem. The LESELAM project falls in line with the ENVALAG and SEDILAG projects, which are led by the university centre and Mayotte Marine Natural Park. They all share a common goal: to raise awareness and to develop tools that will hopefully slow down, in the long run, the siltation of the lagoon.

UNDERSTANDING EROSION IN MAYOTTE

In order to understand how erosion occurs in Mayotte, an observatory was set up in late 2015 as phase one of the LESELAM project. To best represent the island's various environments, three different zones were selected. The first, located on Mtsamboro Hill, is 17 hectares. It represents coastal environments, with steep slopes and dense urbanisation. In Mtsamboro, run-off waters are concentrated in the main gully and flow onto the beach across from the Choazil Islands. The second zone is in Dzoumogne, above the drinking water reservoir along the Oua Bandrani stream. It measures 343 hectares and is representative of environments that have both agricultural land and natural forest areas. Here, there are also a few padzas. Finally, the third zone in Salim Bé is located further south. It combines all three components: natural, agricultural and urban.

WHAT MEASURING TOOLS HAVE BEEN ESTABLISHED?

In Salim Bé, the stream right behind me, measuring tools were set up to measure the water level and water flow. We also measure its turbidity, or the level of suspended matter. We've also set up several rain gauges along the drainage basin, both upstream and downstream, for an accurate measurement of the area's precipitation. 15 to 20 mm of rainfall will result in stormwater runoff that converges along river systems and gullies. This cloudy water is a reddish-orange colour, which is characteristic in Mayotte, and which flows into the lagoon. When the water level rises, our automatic samplers collect water samples from the swollen river. They're analysed in a laboratory to determine the concentration of suspended particles. This is then compared to the turbidity measured on site. Since the end of 2015, we have collected data on erosion, rainfall, and run-off for three drainage basins: Salim Bé, Mtsamboro, and Dzoumogne. They provide interesting data on coastal erosion in tropical environments.

WHAT ARE THE OBTAINED RESULTS?

Mtsamboro's basin went from an erosion rate of 1 ton per hectare per year in 2016 to over 16 tons in the last year. The main cause is the advent of self-built infrastructures in periurban areas. Dzoumogne basin's level of erosion is much lower than Mtsamboro's, due to land use being predominantly natural and agricultural and because of its moderate slopes. However, we have seen a clear evolution between 2016 and 2020. The erosion rate was zero over the first two years and has now reached 0.5 to 1 ton per hectare per year. Mayotte's traditional forests, which provide full land-cover, are being replaced by cassava and banana fields on medium-to-steep slopes. Salim Bé's drainage basin has medium level of erosion at 1.5 to 4 tons per hectare per year due to its intermediate characteristics between Mtsamboro and Dzoumogne. The data collected by LESELAM's observatory allows us to calibrate a watershed erosion model over the three basins, which was applied to all of Mayotte's basins to produce a map of sedimentation flux in the lagoon from all of the island's drainage basins. Mapping helps us prioritize the basins by contribution to the lagoon, identify the more erosive basins, and prioritize measures of erosion control, both in urban and agricultural environments.

WHAT SOLUTIONS FOR CONSTRUCTION SITES EXIST?

At LESELAM's observatory, we have established case-control embankments to measure erosion from the building phase to the final phase. Our measurements went from 300 tons per hectare to 110 tons per hectare the first year and from 50 tons per hectare to 10 tons per hectare the following years after plant preservation measures had been established. We also studied various construction sites on the island where we found a lack of slope stabilization and a lack of consideration towards the fight against erosion in both the conception and completion phases. These measures have allowed us to create a guide for good urban conduct. This guide offers simple measures that are adapted to Mayotte thanks to 10 data sheets that mainly address revegetation, protection of building sites, recommendations for each phase, and monitoring to ensure proper construction practice in Mayotte.

EVOLVING AGRICULTURAL PRACTICES

Along with our partners, we established a Wischmeier system that is made up of two segments: one case-study section and one improved section. In the case-study section, we left the crops in place without changing the way things are done on local farms. In the improved section, we introduced certain practices such as adding pineapple hedgerows, mulch, and cover crops. We tracked the evolution to see, in the rainy season, what forms of erosion we'd find. In the last two rainy seasons, the plots in which we maintained traditional farming methods, methods you might find on any farm, typically with monocultures of cassava and bananas, we measured 30 to 40 tons per hectare of sediments. And plots that had pineapple hedgerows and mulch had less than 1 ton per hectare. This proves the efficiency of these methods. People in Mayotte used to have farms with multiple crops on one plot, which held the soil together. But today, they practice monoculture, planting only one type of crop per plantation. This causes a lot of erosion on the plots. Today, people farm wherever they can, even on slopes. This also leads to a lot of erosion.

WHAT ABOUT SELF-BUILD HOUSING?

"Self-build" is the construction of one's individual home. You are both the contractor and the developer. There are no regulation laws for self-building. However, it is important to respect construction and housing codes and to undertake the necessary steps. To fight against erosion in a self-build project, you may, for example, avoid making terraces during the rainy seasons, protect mounds of earth with tarps, and limit this mound of earth's exposure to the elements. We strongly recommend building an evacuation system for collecting run-off water from plantations and roofs.

GULLIES AND STREAMS: FRAGILE ZONES

Gullies and streams play an important role in erosion. They carry sediments to the lagoon. One must not build a home next to a gully and protect these areas with plant cover. A best-practice guidebook was developed to inform you with the proper conduct to protect these areas.

WHAT OF THE STUDENTS OF MAYOTTE?

In order to teach students about ending erosion, educational resources are made available to teachers on the academy's website. It is also a way of working with digital tools and local data. The data collected for this educational project is the result of a beautiful collaboration between BRGM, particularly members of BRGM who participated in LESELAM, and the island's educational community. This exchange between the scientific and educational communities made this data accessible to a wider population. Students may access this digital data and add to it their local environment. This is one of the themes taught in Year 12: erosion and human activity. Students don't always make the connection with erosion, so it is up to the teachers to raise these questions about the reasons behind this heavy erosion in Mayotte and its consequences. Our students see this erosion but have trouble understanding the causes and consequences of this erosion. One of the main difficulties is being able to project into the future and to see the indirect long-term consequences of erosion on the island of Mayotte, whether in terms of subsistence or agricultural resources. Having quantitative data, for example, allows them to better understand this important phenomenon on the island of Mayotte. It is necessary for students and future generations to become aware of this issue to change our methods and preserve our island.

WHAT NEXT FOR THE LESELAM PROJECT?

LESELAM's observatory will continue monitoring run-off and erosion for the next three rainy seasons as part of phase 3 of the LESELAM project. The conservation techniques that were tested during LESELAM 2 will be deployed on larger-scale sites such as Doujani's periurban site or Mro Mouhou's agricultural site. The goal is for the population, primary schoolers, high schoolers, like those at Coconi Agricultural School, farmers, villagers, or community technicians, to start using these methods themselves so that erosion is no longer inevitable on the island of Mayotte and to limit, as much as possible, the siltation of the lagoon.

BRGM: Geoscience for a Sustainable Earth

Fight against Soil Erosion and Siltation of the Lagoon

Agricultural Outlook for Mayotte in 2035

An Exercise in Participative Forecasting

Among other things, our project is looking at how soil erosion will change in the future.

The answer depends on how we humans use land in the future, how we manage urbanization issues and the type of agriculture we practise there.

To anticipate future changes in agriculture, we decided to implement participative forecasting.

Our goal was to consider the future with local people and to envision how agriculture might change in the long term. We chose 2035 as our horizon. 15 years isn't much, but it is for Mayotte as things change quickly here.

To conduct our exploration of the future, we called upon several groups of people, who worked in tandem. We worked with the state, the regional council, the agriculture department, the land management bureau, and a group of small farmers consisting of a dozen people from our research basin in Mtsamboro in the north.

To help our participants imagine the future, we created three scenarios for 3 possible agricultural transformations. The first imagines current problems continue, which would prevent the emergence of professional, entrepreneurial agriculture. The goal is to make participants think about the result of inaction or what will happen if voluntary policies are not implemented. The other scenarios involve voluntary policies. This lays out actions to enable the development of small family-run agriculture that would add value. The aim is to involve as many locals as possible. The third scenario imagines investor involvement. So more productive companies in limited numbers, which involve fewer farmers, thus leaving more people out in the cold.

The scenarios were presented as fictitious articles describing the imagined changes.

The workshops lasted about three to four hours, with 45 minutes dedicated to each scenario. To foster productive debates, participation was limited to ten people, so all could speak. The workshop ended with a short questionnaire. After discussing each scenario, they answered questions, allowing us to quantify their final positions.

To work with small farmers, our approach had to be adapted as our method was based upon articles written in French. This was not ideal for the farmers, some of whom were illiterate. So the scenarios were summarized orally. The farmers fully understood the scenarios and the related hypotheses. Then we asked them to tell us how the scenarios could be adapted to their farms and for their farmers' group. For this, we used simplified representations of farms. We presented the example of a collective farm, as proposed in one of the scenarios, and we asked them how they would organize collectively for a project of this type. The discussion was rich and lively. Some people in the group espoused a traditional approach while others took a more innovative approach. The workshops brought these ideas to the fore. Interesting conclusions were drawn. This way of bringing farmers together to reflect was new for them. And the fact is, there's a long way to go before a project emerges from within the group.

To conclude, we organized a follow-up session with all of the participants.

This session allowed the groups to share their conclusions. We saw that there was little disagreement among them and that they had moved towards a shared vision of the agriculture they wished to develop by 2035. The workshops provoked reflection on the future, but there's much to do. Other people will take over now, to define an agricultural development strategy as well as envision and plan concrete actions to achieve the scenarios that we discussed.