Executive Summary: In the current scenario of global change, some of its effects are much less known than others. Despite the fact that we all accept that, as a result of such a change, the temperature will increase and the sea level will rise, for few it is unthinkable that our ecosystems (for example, the cloud forest or the Andean páramo) are also being affected.
The effect of global change on these ecosystems is such that the border between them is "moving" and this movement is in the direction of consumption (ie loss) of the Ecuadorian páramo. However, there are still no robust models that relate climate variables with the magnitude and direction of ecosystem movement. In this sense, the present project intends to carry out the modeling of the ecosystem border through the use of optimal and robust signal processing techniques. The experiment will be carried out at the Yanahurco Hacienda, on the slopes of the Cotopaxi volcano, right on the border with the beginning of the Napo; being the part corresponding to the Cotopaxi páramo and the part corresponding to the beginning of the Napo montane forest. The coordinates of the project are: -0.692273; -78.247119 and -0.68333; -78.23302. On this border, a distributed sensor network will be located and, through an Internet connection infrastructure powered by solar energy, the data stream from the sensors will be sent to the cloud for real-time viewing and, through an interface ( front-end, backend), users can have the desired information. In this project, the ecological, biophysical and climatic variables of interest will be identified. Then, characteristics of the population will be obtained from samples extracted from it and the distribution of said population will be estimated, carrying out robust data treatment. Next, mathematical models will be obtained that allow us to describe the dynamics of the processes under study and signal processing techniques will be applied for the filtering and estimation of variables.
Likewise, computational modeling tools (such as: support vector machines, among others) will also be used to carry out the identification of the dynamics of the páramo-forest border and of the biomass activity in it.
General Objective: Robustly characterize the páramo-forest ecosystem boundary in the Ecuadorian Andes.
Specific objectives:
- Install a network of sensors in the region under study that allows us to monitor in real time the behavior of the variables of interest under measurement.
- Design an interface that is responsible for collecting the data and transforming it, for its subsequent processing and delivery to the user. This interface consists of a frontend and a backend.
- Modeling of the ecosystem frontier:
- Identification of ecological, biophysical and climatic variables, among others.
- Collection/Generation of biophysical variables for the study area (altitude, slope, geoform, coverage).
- Obtaining population characteristics from samples taken from it and estimating the distribution of said population.
- Robust data processing and modeling.
- Optimal signal processing and estimation of variables.
- Identification and modeling of ecosystem boundary dynamics using artificial intelligence tools, such as: support vector machines, data mining, etc.
- Identification of the biomass activity dynamics in the páramo-forest frontier using data from remote sensors (since the year 2000), and identification of cover changes from 1980 to 2000, taking two tentative periods.
Participating Institutions:
UTPL, UDLA, PUCESI.
Participants:
Project Director Wilmar Hernández Perdomo
- Wilmar Hernandez Perdomo
- Fabian René Reyes Good
- Aminael Sanchez Rodriguez
- Yunierkis Perez Castillo
- Laura Guerra Torrealba
- Sweet Rivero
- Stalin Arciniegas
- Luis D. Narvaez
- Santiago Quishpe
- frankiln rivas
Awarded budget: $26250
Project status: Closed early.
