MODELING SPATIAL PATTERN OF THE EFFECTS OF HEAT ON LAYING BIRDS' PRODUCTIVITY IN SOUTH WESTERN NIGERIA

Study of the spatial analysis of the spatial variability on the effects of weather indices on layers’ productivity have beennecessitated by the global climatic changes which consequently lead to weather instability. A study of the spatial analysis of the effects of some selected weather indices on layers’ productivity wasconducted using data collected from 5 randomly selected farms from the study area. The objective of the study was to investigate the spatial variation of the effects of the selected weather indices onlayers productivity. The results of the descriptive analysis showed that both sample statistics (mean and variance) were bias estimates of the selected variables. The analysis of variance revealed thatthere were significant differences in the means returned by each of the variables for the different sites. The F statistics, 31.39,31.57, 125.06, 4.05 and 1059.56 returned concurrently for henday production, rectal temperature, room temperature, relativehumidity and temperature humidity index were greater than F (4,295:0.01) = 3.32. The result of the mixed model analysis showedthat no 2 covariance structure gives the same estimation though the estimations may be fairly close. Similarly, it was revealed that site 5’s contribution to the spatial description of the effects of weather on layers’ productivity is null and the interaction of day by site 5 was null also. The implication of this is that site 5 has no contribution(s) in the description of the effects of weather on layers productivity of the study area. The information criteriaindicated that both Unstructured and Autoregressive returned closely the least of Akaike Information Criteria (AIC) or Corrected Akaike Information Criteria (AICC). However, autoregressive covariance structure gave the least AIC (3053.7), AICC (3053.80) and BayesianInformation Criteria, BIC (3053). HuynhFeldt (4531.8) and unstructured (7471.0) covariance structure gave the highestinformation criteria.