Field monitoring, leaching studies, and experimentation in soil biology are often now being done non-destructively using Æxed installations so that measurements are made repeatedly on thesameunits. Theresulting data for each unit (suction cup, lysimeter, incubation chamber) constitute a time series in whichthere may be autocorrelation. The usual methods of statistical analysis, such as the analysis of variance,must be modiÆed or replaced by more suitable ones to take account of the possible correlation.This paper describes the split-plot design of such experiments, shows how to assess the variance±covariance matrix of residuals for uniformity by the Greenhouse±Geisser statistic, and shows how to usethis statistic to adjust the degrees of freedom in a formal test of signiÆcance. It also describes more recentmethods. Ante-dependence analysis identiÆes the extent of the temporal correlation in the data andprovides approximate signiÆcance tests for the treatments. Alternatively, the paper also shows how thetraditional analysis of variance may be replaced by a restricted maximum likelihood analysis which givesWald statistics.The techniques are illustrated with data on CO2evolved from soil incubated for 75 days in closedchambers, during which time the gas was measured on 24 occasions to give time series for threereplicates of each combination of two soils (limed and unlimed) and three types of ryegrass amendment.An ante-dependence structure (extending to ninth order) weakened the usual signiÆcance test within thesubunit stratum. The Wald statistics showed that there was, nevertheless, a strong interaction between thetreatments and time.
{{menu_about_desc}}
Explore eventos futuros e passados em todo o mundo e online, sejam hospedados pelo CIFOR-ICRAF ou com a participação de nossos pesquisadores.
Découvrez les évènements passés et à venir dans le monde entier et en ligne, qu’ils soient organisés par le CIFOR-ICRAF ou auxquels participent nos chercheurs.
Jelajahi acara-acara mendatang dan yang telah lalu di lintas global dan daring, baik itu diselenggarakan oleh CIFOR-ICRAF atau dihadiri para peneliti kami.
Explore eventos futuros y pasados organizados por CIFOR-ICRAF o con la participación de nuestros investigadores.
{{menu_nowledge_desc}}.
{{menu_research_desc}}
{{menu_about_desc}}
CIFOR–ICRAF achieves science-driven impact. We conduct innovative research, strengthen partners’ capacity and actively engage in dialogue with all stakeholders, bringing the latest insights on forests, trees, landscapes and people to global decision making.
CIFOR–ICRAF publishes over 750 publications every year on agroforestry, forests and climate change, landscape restoration, rights, forest policy and much more – in multiple languages.
Explore upcoming and past events across the globe and online, whether hosted by CIFOR-ICRAF or attended by our researchers.
CIFOR–ICRAF addresses local challenges and opportunities while providing solutions to global problems for forests, landscapes, people and the planet.
We deliver actionable evidence and solutions to transform how land is used and how food is produced: conserving and restoring ecosystems, responding to the global climate, malnutrition, biodiversity and desertification crises. In short, improving people’s lives.
