Table 1.
Manure storage periods, including start dates, partial pump-out dates, complete pump-out dates, and the seasons they cover.
Fig 1.
(a)The daily mean ambient air temperature, AAT; (b) manure temperatures at the different locations along the depth axis.
The vertical dashed lines represent when total pump-outs occurred. The dotted lines indicate partial pump-out events. The storage periods are labeled I – VII.
Fig 2.
(a) Daily mean ambient air temperature, AAT; and (b) daily mean manure temperature at various locations along the depth axis showing a thermal turnover event during a cold storage period.
Fig 3.
(a) Daily mean ambient air temperature, AAT; (b) daily mean manure temperature at various locations along the depth axis showing a thermal turnover event during a warm storage period.
Table 2.
Mean manure temperatures at various depth locations and ambient air temperatures (± SE) during each storage period (I–VII). Temperatures within each column with the same letter are not significant. The average temperatures within each row (Period) with the same symbol are not significantly different (Tukey HSD, p < 0.05).
Fig 4.
Distribution of manure temperatures at various depth locations (045 = 0.45 m, 090 = 0.90 m, 168 = 1.68 m, and 244 = 2.44 m above the pit bottom) and ambient air temperature (AAT) across storage periods I–VII.Boxes represent the interquartile range (25th–75th percentile), with the median shown by the horizontal line.
Fig 5.
Correlation heatmaps showing relationships between meteorological factors and manure temperature at locations 045 (0.45m), 090 (0.90 m), 168 (1.68 m), and 244 (2.44 m) above the bottom of the storage pit. Color gradients indicate the strength and direction of correlations, with red denoting positive and blue negative. Panel (a) shows Pearson correlation coefficients (linear relationships), and panel (b) shows Spearman rank coefficients (monotonic relationships). Statistical significance is indicated by asterisks (* p < 0.05, ** p < 0.01, *** p < 0.001). Meteorological variables include ambient air temperature (AAT), wind speed (WS), wind direction components (WD_sin and WD_cos), solar radiation (SR), relative humidity (RH), and rainfall (RF).
Table 3.
Standardized regression coefficients for meteorological factors of manure temperature across measurement locations along the depth axis.
Table 4.
Variance decomposition of meteorological predictors explaining manure temperature across four locations.
Table 5.
Model fit statistics for models evaluating the moderating effect of manure volume on the relationship between meteorological predictors and manure temperature across four locations.
Table 6.
Performance comparison (R2) between standard Linear Regression and non-linear Gradient Boosting Regressor of manure temperature across locations.
Fig 6.
Permutation importance from Gradient Boosting Regressor models predicting manure temperature at locations 045, 090, 168, and 244 along the depth axis.
Meteorological factors predictors include ambient air temperature (AAT), wind speed (WS), wind direction (WD_sin and WD_cos), solar radiation (SR), relative humidity (RH), rainfall (RF), and manure depth (dm; proxy for manure volume).