Identifying and Removing Spatial Correlation from Yield Experiments

Abstract

In classical statistics, the effect of soil trends is compensated for by replication and randomization of treatments. Two field experiments were conducted at sites with significant soil trends to evaluate the use of semivariograms for identifying spatial correlation in plot yield, and evaluate the ability of nearest‐neighbor analysis (NNA) in removing trend. The first experiment involved a P‐fertilizer trial with winter wheat (Triticum aestivum L.) on an eroded hillslope in eastern Washington. The second experiment involved a N‐ and P‐fertilizer trial with cotton (Gossypium hirsutum L.) in Dera Ismail Khan, Pakistan. One of the difficulties in using semivariograms of plot‐yield data to evaluate spatial correlation in experimental errors is that yield is affected by the presence of trends as well as by the pattern in treatment randomization and replication. To remove the influence of treatment randomization, the measured mean for each treatment was subtracted from the measured yield for that treatment in each plot. Semivariograms of these deviations in yield relative to the treatment mean showed significant structure for both experiments, indicating spatial correlation between plots resulting from soil trends. We used NNA to adjust measured plot yields for the effects of spatial correlation. Semivariograms of yield deviations after this adjustment exhibited no spatial structure, indicating removal of spatial correlation between plots. Analysis of variance (ANOVA) on measured yields before adjustment by NNA in both experiments showed nonsignificant treatment effects, while block effects were highly significant. Thus, without adjustment, the experimental results showed no response to the applied fertilizer treatments. In contrast, ANOVA on adjusted yields after NNA showed highly significant treatment effects, while block effects were nonsignificant.

Delineation of Management Zones Using Soil Properties and a Multivariate Geostatistical Approach

“…Determination of these sub-field areas is difficult due to the interactions among several biotic, abiotic and climate factors that affect crop yield. Different approaches have been developed for site-specific management zone delineation: topography and soil properties, including soil survey maps (Carr et al, 1991) and soil sampling (Mulla, 1991); landscape factors obtained from digital elevation models (DEM) (McCann et al, 1996;Lark, 1998;Nolan et al, 2000), aerial photographs of bare soil (Fleming et al, 2000), remote sensing images (Bhatti et al, 1991;Moulin et al, 1998), invasive soil sampling (Mulla, 1991) and non-invasive soil sampling using electrical conductivity Delineation of Management Zones Using Soil Properties and a Multivariate Geostatistical Approach (Johnson et al, 2003;Castrignanò et al, 2006). Therefore, combining several types of data (soil, landscape, crop, yield and multiple remote sensing images) is supposed to partly compensate the weakness of the approaches using a single source of information and, above all, to produce more stable management zones.…”

Efficiency of an Incomplete Block Design Based on Geostatistics for Tillage Experiments

“…Likewise, when local 7 heterogeneity is spatially structured, randomization and replication alone do not ensure that all 8 treatment comparisons are made with equal precision (van Es and van Es, 1993). Thus, spatial 9 dependence of observations has been cited as a major reason for nonsignificant treatment 10 differences using a classical complete block analysis in tillage and yield experiments (Roseberg and 11 McCoy, 1988;Bhatti et al, 1991;Ball et al, 1993). Control of these effects of soil variability is of 12 special interest in long-term experiments examining treatments that are generally expected to have 13 relatively moderate and slowly developing effects (Frye and Thomas, 1991).…”

Heterogeneity, plot shape effect and optimum plot size

“…Results along these lines are given by Starks ( 1986 ) and Zhang et al (1990). Crop heterogeneity has also been studied using geostatistical techniques by Bhatti et al (1991).…”