Multi-species allometric models to predict biomass in tropical high Andean grasslands
The models of growth and allometry of grasslands allow to know growth and development phenomena. Consequently, the aim of this study was to develop a multi-species allometric model to estimate biomass in three grasses in the High Andean Zone of Jauja, Peru, at an altitude of 3600 m above sea level. The native grasses were Festuca dolichophylla (J. Presl) tall (50-90 cm), Festuca humilior (Nees & Meyen) medium (20-35 cm) and Calamagrostis vicugnarum (Weddell) Pilger short (5- 25 cm). The measured variables were biomass (g, YB), basal diameter (cm, XD), canopy diameter (cm, XC) and height (cm, XH). The potential regression was used by choosing the strength of the models by the Akaike criterion (IAC) and the adjusted coefficient of determination (adjusted R2). The multispecies allometric model was Y = 1,238 * XD2,231, due to its lower IAC (383.795) and higher adjusted R2 (85.4%). C. vicugnarum (Weddell) Pilger had an effect on the multi-species model based on the basal diameter. The multi-species model in F. dolichophylla (J. Presl) shows average estimates at 37% below the individual model. The explanatory variable of basal diameter has greater modelling strength for native grass grasses. It should also be considered that the height of the flag leaf had a greater strength than the diameter of the canopy. The multispecies allometric models allowed to increase the predictive strength in species of high, medium and low size; however, the behaviour of the slope in C. vicugnarum (Weddell) Pilger and the multispecies and individual model relationship in F. dolichophylla (J. Presl) must be considered.
| Main Authors: | , , , , , |
|---|---|
| Format: | Digital revista |
| Language: | spa |
| Published: |
Universidad Nacional Mayor de San Marcos, Facultad de Medicina Veterinaria
2020
|
| Online Access: | https://revistasinvestigacion.unmsm.edu.pe/index.php/veterinaria/article/view/19034 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|