Limits...
A Four Dimensional Spatio-Temporal Analysis of an Agricultural Dataset.

Donald MR, Mengersen KL, Young RR - PLoS ONE (2015)

Bottom Line: The proposed models vary in their representation of the spatial correlation in the data, the assumed temporal pattern and choice of conditional autoregressive (CAR) and other priors.In terms of the substantive question, we find that response cropping is generally more effective than long fallow cropping in reducing soil moisture at the depths considered (100 cm to 220 cm).Thus, if we wish to reduce the possibility of deep drainage and increased groundwater salinity, the recommended cropping system is response cropping.

View Article: PubMed Central - PubMed

Affiliation: Mathematics and Statistics, University of New South Wales, Sydney, NSW, Australia.

ABSTRACT
While a variety of statistical models now exist for the spatio-temporal analysis of two-dimensional (surface) data collected over time, there are few published examples of analogous models for the spatial analysis of data taken over four dimensions: latitude, longitude, height or depth, and time. When taking account of the autocorrelation of data within and between dimensions, the notion of closeness often differs for each of the dimensions. Here, we consider a number of approaches to the analysis of such a dataset, which arises from an agricultural experiment exploring the impact of different cropping systems on soil moisture. The proposed models vary in their representation of the spatial correlation in the data, the assumed temporal pattern and choice of conditional autoregressive (CAR) and other priors. In terms of the substantive question, we find that response cropping is generally more effective than long fallow cropping in reducing soil moisture at the depths considered (100 cm to 220 cm). Thus, if we wish to reduce the possibility of deep drainage and increased groundwater salinity, the recommended cropping system is response cropping.

No MeSH data available.


Method 1: Spatially structured and unstructured standard deviations.Top panel: and 95% credible intervals at depths 100 cm. Bottom panel: 95% credible intervals at depths 220 cm. The spatial standard deviations are shown using heavy black lines and bars, the unstructured standard deviations using dotted lines and bars.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4626095&req=5

pone.0141120.g001: Method 1: Spatially structured and unstructured standard deviations.Top panel: and 95% credible intervals at depths 100 cm. Bottom panel: 95% credible intervals at depths 220 cm. The spatial standard deviations are shown using heavy black lines and bars, the unstructured standard deviations using dotted lines and bars.

Mentions: Not unexpectedly, the variances at the shallower depths show greater variability across the sampling dates, than those at greater depths. Fig 1 illustrates this, showing the square roots of the spatial and the unstructured variances for each date at depths 100 cm and 220 cm based on the model of Method 1. The comparable graphs across all depths show decreasing variability with increasing depth of these parameters across the sampling dates (Figs Y-AE in S1 File). This decreasing variability with depth is also seen in the contour smooth for the variance components across days and depth shown in the two lower panels of Fig 2. The variability in these parameters justifies the choice to fit the same model across all sampling dates thereby allowing the parameters of the original model to vary by date, since a description of their evolution across time was not obvious a priori.


A Four Dimensional Spatio-Temporal Analysis of an Agricultural Dataset.

Donald MR, Mengersen KL, Young RR - PLoS ONE (2015)

Method 1: Spatially structured and unstructured standard deviations.Top panel: and 95% credible intervals at depths 100 cm. Bottom panel: 95% credible intervals at depths 220 cm. The spatial standard deviations are shown using heavy black lines and bars, the unstructured standard deviations using dotted lines and bars.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4626095&req=5

pone.0141120.g001: Method 1: Spatially structured and unstructured standard deviations.Top panel: and 95% credible intervals at depths 100 cm. Bottom panel: 95% credible intervals at depths 220 cm. The spatial standard deviations are shown using heavy black lines and bars, the unstructured standard deviations using dotted lines and bars.
Mentions: Not unexpectedly, the variances at the shallower depths show greater variability across the sampling dates, than those at greater depths. Fig 1 illustrates this, showing the square roots of the spatial and the unstructured variances for each date at depths 100 cm and 220 cm based on the model of Method 1. The comparable graphs across all depths show decreasing variability with increasing depth of these parameters across the sampling dates (Figs Y-AE in S1 File). This decreasing variability with depth is also seen in the contour smooth for the variance components across days and depth shown in the two lower panels of Fig 2. The variability in these parameters justifies the choice to fit the same model across all sampling dates thereby allowing the parameters of the original model to vary by date, since a description of their evolution across time was not obvious a priori.

Bottom Line: The proposed models vary in their representation of the spatial correlation in the data, the assumed temporal pattern and choice of conditional autoregressive (CAR) and other priors.In terms of the substantive question, we find that response cropping is generally more effective than long fallow cropping in reducing soil moisture at the depths considered (100 cm to 220 cm).Thus, if we wish to reduce the possibility of deep drainage and increased groundwater salinity, the recommended cropping system is response cropping.

View Article: PubMed Central - PubMed

Affiliation: Mathematics and Statistics, University of New South Wales, Sydney, NSW, Australia.

ABSTRACT
While a variety of statistical models now exist for the spatio-temporal analysis of two-dimensional (surface) data collected over time, there are few published examples of analogous models for the spatial analysis of data taken over four dimensions: latitude, longitude, height or depth, and time. When taking account of the autocorrelation of data within and between dimensions, the notion of closeness often differs for each of the dimensions. Here, we consider a number of approaches to the analysis of such a dataset, which arises from an agricultural experiment exploring the impact of different cropping systems on soil moisture. The proposed models vary in their representation of the spatial correlation in the data, the assumed temporal pattern and choice of conditional autoregressive (CAR) and other priors. In terms of the substantive question, we find that response cropping is generally more effective than long fallow cropping in reducing soil moisture at the depths considered (100 cm to 220 cm). Thus, if we wish to reduce the possibility of deep drainage and increased groundwater salinity, the recommended cropping system is response cropping.

No MeSH data available.