Angga Yudaputra
Through the USAID Forestry Fellowship program, I earned a master’s degree from the School of Forest Resources and Conservation at the University of Florida. I am grateful I earned my degree and I really benefited from my supervisor who is an expert in the field of ecosystem simulation and modeling.
When I started my master’s, I was confident I would continue to work with the software for plant conservation purposes that I was already familiar with, but my supervisor suggested I try machine learning instead. Although it was stressful, I followed his advice and I learned a lot.
I studied programming languages like R and Python through modelling courses, online tutorials, books and active involvement in online programming discussion threads. It has been two years since I graduated and I am working on modelling and simulation to understand the impact of global changes, including climate change, natural disasters, natural fire, and the consequences of human land-use change and unsustainable harvesting on the native and threatened plants of Indonesia. I now use several approaches, combining computer programming languages, geographic information systems (GIS) and remote sensing to simulate habitat preferences and population.
The conservation of individual species should be based on knowledge of habitat requirements and the population demographic status. It is impractical to assess each of thousands of species in ecoregion, but rare or narrowly distributed species may constitute important conservation targets. Knowledge of habitat requirements is quite important to design conservation areas for these species and to promote long-term persistence (both through in-situ or ex-situ conservation). Additionally, knowledge of population structure and demography is needed to assess viability of populations. In this study, I investigate habitat suitability and population size structure for a newly identified endemic palm species (Pinanga arinasae) The palm has an important role in the indigenous human community. Plots with palms and adjacent areas with no palms were sampled to characterize key habitat variables. Habitat suitability was modeled using machine learning techniques of Artificial Neural Network (ANN) and Random Forest (RF). Population size structure was characterized by counting and measuring the height of individuals found in plots. The ANN variables that best explain occurrence were litter depth, elevation, canopy openness and slope. The RF variables that best explained the data were elevation, litter depth, slope, and aspect. Both the Artificial Neural Network 11 (ANN) and Random Forest (RF) are robust models that can be used to predict the occurrence of Pinanga arinasae. The population size structure showed that there are many seedlings, but juvenile and mature individuals were found in small numbers.
Partners
