Analysis of spatial and temporal variability of maize productivity based on physical parameters

Mestrado em Engenharia Agronómica / Instituto Superior de Agronomia. Universidade de LisboaThe evaluation of a field´s spatial and temporal variability, together with the mapping of its characteristics, is an important tool to help decision-making. That said, this research will have the following objectives: 1) Identification of variable and non-variable fields. 2) Study which physical parameters drive time variations in average yields. 3) Study which physical parameters drive spatial variations in average yields and to what extent the variability zones have different yields and different responses to the factors under study. 4) Development of spatial trend maps based on the combination of spatial and temporal variability in two fields of study. To meet the proposed objectives, data, referring to the years 2015-2019, were used, provided by Quinta da Cholda, Azinhaga, Portugal. The identification of variable and non-variable fields was made according to the temporal variability measured by the standard deviation. The results showed that only 5 out of 13 fields presented great income inconsistency in the defined period. For the study of temporal and spatial variation´s driving factors, multiple linear models were created. Regarding the temporal variation, the best model revealed to be the model integrating the interactions between the apparent soil electrical conductivity, the topographic wetness index, altimetry, together with the addition of the variable soil type, obtaining a model capable of explaining 81% of the existing variability. As for spatial variation, measured by average yields, two approaches were used. The first resulted in a model with the same structure as the previous one mentioned, obtaining a model capable of explaining 54% of the existing variability and the second resulted in an additive model capable of explaining 96% of the existing variability. The development of a spatial trend map allowed to create a map, which combines spatial and temporal variability and may be used in the future for the design of study and intervention zones.N/