Ecological Indicators · 2020 · 16 citations · 56 references
Natural ResourcesConservation BiologyEcologyFish BiologyFish EcologyFreshwater EcologyFish Habitat AssessmentFreshwater BiodiversityFreshwater Biodiversity ConservationFreshwater Ecosystem EcologyFreshwater Fish ConservationFreshwater Fish EcologyEndangered Freshwater FishesFish Distribution PatternsFish Habitat ModelingHabitat Suitability MappingSpecies Distribution ModellingSpecies Distribution PatternsThermal RegimesThreatened Species
The key challenge in the fields of conservation and resource management is to track local, regional and global species extinctions and preserve biodiversity through optimization of investment of funds. To fulfil that purpose modelling and mapping habitat suitability of species is becoming essential. The freshwater ecosystem is today one of the most threatened ecosystems and fishes are especially vulnerable for climate change as they have fewer choices for dispersal. The altered thermal regimes, land use and hydrology together cause isolation, fragmentation and changes in distributional patterns of fish. In such landlocked situation, there is utmost need to identify suitable environmental conditions for a species to exist. Indigenous fish is one of the essential components of healthy inland aquatic ecosystems as they fulfil several important ecological functions and form an important part of the aquatic food web. Amblypharyngodon mola is an economically and ecologically important indigenous freshwater fish species of the tropical and subtropical floodplain rivers and wetland habitats of India and Bangladesh. However, lack of knowledge about the key environmental indicators is hindering appropriate management and conservation initiation. Modelling and mapping the distribution of A. mola was done for identifying the key environmental variables and predicting the potential distribution area of the species in current climatic condition. The species occurrence data were accessed from both primary and secondary sources and used after cleaning and removing biasness. Climatic, topographic, landcover and soil variables of 1 km resolution were used as environmental predictors. India and Bangladesh together were used as model background. Ensemble species distribution modelling was done, and the model was validated by area under the curve (AUC) value (>0.7). The result showed that the Ganges delta of lower reach going through West Bengal and Bangladesh is most suitable for A. mola occurrence. Annual upstream precipitation (hydroclim_12), soil pH in water (soil_average_02) and elevation (ele_01) are the key indicators for predicting the habitats suggesting the importance of hydroclimatic, topographic and soil variables in determining the distribution of the fish.
56
Leo Breiman · Machine Learning · 2001 · 126.2K citations · Full text
Classification and Regression Trees.
John Van Ryzin, Leo Breiman, Jerome H. Friedman et al. · Journal of the American Statistical Association · 1986 · 21K citations
Classification and regression trees
J. Praagman · European Journal of Operational Research · 1985 · 10.2K citations
Multivariate Adaptive Regression Splines
Jerome H. Friedman · The Annals of Statistics · 1991 · 8.1K citations · Full text