Estimation of forage yields in Burkina Faso's climatic zones using satellite data
DOI:
https://doi.org/10.19182/remvt.37009Keywords
Pastures, forage yield, above ground biomass, agroclimatic zones, satellite imagery, linear models, Burkina FasoAbstract
The assessment of forage resources is a key element for governing livestock food crises in Burkina Faso. This study aims to evaluate, for the first time, the possibility of estimating forage yields in Burkina Faso’s climatic zones using uni and multivariate linear statistical models constructed from forage plant biomass data collected in the field in 2017, 2018, and 2019, and phenological satellite variables (Normalized Difference Vegetation Index (NDVI) and Fraction of Absorbed Photosynthetically Active Radiation (FAPAR)) and agroclimatic variables (precipitation, soil moisture, evapotranspiration, surface temperature). An exhaustive search for the best linear statistical models with one to four variables was conducted, and the best models according to the Bayesian Information Criterion were identified. The performance of the uni to quadri-variate models obtained is quite low, with for all climatic zones except the Sahelian zone, RRMSE press ranging from 55% to 61% (R² press ranging from 0.07 to 0.36), and for the Sahelian climatic zone, RRMSE press ranging from 42% to 49% (R² press ranging from 0.59 to 0.69). The decrease in correlation of the majority of variables with forage plant biomass along the North-South gradient results in a decrease in model performance along this gradient. Agroclimatic variables were found to be useless, and those derived from FAPAR appeared to be generally more effective than those derived from NDVI. The results also show a very low added value of multivariate models compared to univariate models, except for the Sahelian zone, and a better performance of models developed in more homogeneous climatic zones. A series of recommendations have been identified to improve the coupling between field-collected forage plant biomass data and variables extracted from satellite images, and thereby improve the performance of the models.
Downloads
Downloads
Additional Files
-
Abstract407
-
pdf (Français)122
-
Matériel supplémentaire / Supplementary Material (Français)9
Received
Accepted
Published
How to Cite
Issue
Section
Categories
License
© W.Some et al., hosted by CIRAD 2024
This work is licensed under a Creative Commons Attribution 4.0 International License.