Distribution, ecological and economic impacts and competition of the invasive alien aquatic weeds (Pontederia crassipes Mart., Pistia stratiotes L., Salvinia molesta D.S. Mitch. and Azolla filiculoides Lam.) in Madagascar

In Madagascar, as in several countries in the world, the invasion by four aquatic weeds (Pontederia crassipes Mart. (Pontederiaceae), Pistia stratiotes L. (Araceae), Salvinia molesta D.S. Mitch Salviniaceae) and Azolla filiculoides Lam. (Azollaceae) are among the drivers of environmental and socio-economic deterioration in aquatic ecosystems. Pistia stratiotes was first recorded on the island in the 19th century, and P. crassipes from the beginning of the 20th century, while S. molesta and A. filiculoides were only documented during in the 21st century. From the 1920s, botanists such as Henri Perrier de la Bathie and Raymond Decary were already aware of the dangers caused, in particular by P. crassipes in other countries, and raised the alarm, but little attention has been paid to these species. The aim of the research conducted for this thesis was to determine the distribution, socio-economic and ecological impacts of these four invasive alien aquatic weeds in Madagascar and to make recommendations for their control. First, the distributions of these four aquatic weeds were mapped. This mapping exercise compiled data from different sources including herbarium records, online data and field visits across Madagascar. The mapping study was undertaken from August 2015 to June 2020. Except for mountainous areas above 1800 m (Tsaratanana Massif, Ankaratra Massif and Andringitra Massif) where no data were available, all of Madagascar's bioclimates were invaded by at least one of the four aquatic weeds. In total, at least one species was recorded in 18 of the 22 Regions. Pontederia crassipes was recorded in 13 Regions, S. molesta in 14 Regions, P. stratiotes in 12 Regions, and A. filiculoides in 13 Regions. Herbarium records revealed the oldest record for P. stratiotes to be 1847, 1931 for P. crassipes, 1995 for S. molesta and there were no herbarium specimens for A. filiculoides prior to the start of the current study in 2015. We now know where these four weeds occur and how abundant they are. An objective of this research was to assess the impacts of the four invasive aquatic plants on the socio-economy of the island, mainly on rice production and fishing. Between 2016 and 2019, 102 households in three regions, Soanierana Ivongo, Foulpointe and Antananarivo, were randomly selected and questioned on the impact of these weeds in their aquatic ecosystems and their livelihoods such as fishing and rice growing. Surveys revealed that the four aquatic weeds significantly threatened household activities. On the east coast of Madagascar, the invasions of these four invasive species decreased fish and freshwater shrimp production by 82%. On the high plateau of Madagascar, they reduced rice yield by 30% despite requiring an additional expense of US$ 1,107/ha for control. Although farmers surveyed only used manual control to manage these weeds, they were receptive to other control methods, including integrated control using herbicides and biological control. Another objective of this research was to determine the ecological impacts of the four weeds and specifically if freshwater ecosystem functioning would return after control. To assess the ecological impact, between February 2017 to August 2019, on Lake Antsokafina, the following abiotic and biotic factors were considered: physico-chemistry of water, succession of macrophyte community and animal diversity. With the exception of turbidity, the values of the physico-chemical parameters of the water (pH, electrical conductivity, water temperature and turbidity), were similar between the infested zone and cleared zone. A study on the invasion process of aquatic weeds showed that the plant community succession of the lake changed over time in the areas that had been cleared. The submerged species Ceratophyllum demersum was the pioneer, followed by creeping species such as Echinochloa colona and Ipomoea aquatica, before the area was recolonized by aquatic weeds. Among the aquatic weeds, S. molesta was the most aggressive, covering 92% of the area one year after the start of the experiment. For animal diversity, bird, shrimp and fish community were assessed. The cleaning of the plots in the lake allowed the resumption of fishing activity providing 50 to 200g/catch for shrimp and from 0.25 to 0.5kg/catch for fish per person per day, while no catch was obtained in the areas infested by aquatic weeds were fishermen still attempting to harvest fish/shrimp from the aquatic weed infested areas. Three species of birds, Humblot’s Heron (Ardea humbloti), the white-faced whistling duck (Dendrocygna viduata) and red-billed teal (Anas erythrorhyncha) returned once the areas had been cleared. A manipulated outdoor as descriptor for laboratory experiment was conducted to determine the level and nature of competition of four aquatic weeds species against the indigenous floating fern, Salvinia hastata Desv. (Salviniaceae), using an additive series density model. It was shown that all four invasive species outcompeted S. hastata, with P. crassipes being 24 times more dominant, followed by P. stratiotes at 12 times, S. molesta at 8 times, and finally A. filiculoides at 1.2 times more dominant. This study provided direct evidence of the biodiversity impact of these four species and thus also provided an environmental argument for their control. Based on the findings of this study, a series of recommendations was formulated to manage the invasions of alien species in Madagascar with particular attention to invasive aquatic weeds. These recommendations mainly concern the establishment of management structures and legal instruments such as the creation of a lead government agency at national level and a cross-sectorial invasive species advisory committee, which should review legislation and regulations related to invasive species.