Classification and multivariate analysis of differences in gross primary production at different elevations using biome-bgc in the páramos, ecuadorian andean region
Guardado en:
| Autores: | , , , , |
|---|---|
| Formato: | artículo original |
| Estado: | Versión publicada |
| Fecha de Publicación: | 2015 |
| Descripción: | Gross primary production (GPP) in climate change studies with multi- species and elevation variables are difficult to measure and simulate. Models tend to provide a representation of dynamic process through long-term analysis by using generalized parameterizations. Even, current approaches of modelling do not contemplate easily the variation of GPP at different elevations for different vegetation types in regions like páramos, mainly due to data unavailability. In these models information from cells is commonly averaged, and therefore average elevation, ecophysiology of vegetation, as well as other parameters is generalized. The vegetation model BIOME- BGC was applied to the Ecuadorian Andean region for elevations greater than 4000 masl with the presence of typical vegetation of páramo for 10 years of simulation (period 2000-2009). An estimation of the difference of GPP obtained using a generalized altitude and predominant type of vegetation could lead to a better estimation of the uncertainty in the magnitude of the errors in global climate models. This research explores GPP from 3 different altitudes and 3 vegetation types against 2 main climate drivers (Short Wave Radiation and Vapor Pressure Deficit). Since it is important to measure the possible errors or difference in the use of averaged meteorological and ecophysiological data, here we present a multivariate analysis of the dynamic difference of GPP in time, relative to an altitude and type of vegetation. A copula multivariable model allows us to identify and classify the changes in GPP per type of vegetation and altitude. The Frank copula model of joint distributions was our best fit between GPP and climate drivers and it allowed us to understand better the dependency of the variables. These results can explore extreme situations where averaged simplified approaches could mislead. The change of GPP over time is essential for future climate scenarios of the ecosystem storage and release of carbon to the atmosphere. Our findings suggest that a classification of the difference is highly important to be extended to cells that have similar properties. |
| País: | Portal de Revistas UCR |
| Institución: | Universidad de Costa Rica |
| Repositorio: | Portal de Revistas UCR |
| Lenguaje: | Inglés |
| OAI Identifier: | oai:portal.ucr.ac.cr:article/21602 |
| Acceso en línea: | https://revistas.ucr.ac.cr/index.php/matematica/article/view/21602 |
| Palabra clave: | multivariate classification copula BIOME-BGC GPP páramos clasificación multivariada cópula |