Review: critical appraisal of methods to investigate the effect of temperature on grapevine berry composition

The grapevine is an economically important crop and a model species of long-lived perennials widely used in bioclimatic studies. Ambient temperature modulates berry composition, hence the significant research effort in establishing links between temperature, berry composition and wine attributes. Our current understanding of the effect of temperature on berries and wines has been largely gained from indirect methods or direct methods in controlled conditions. Indirect methods include comparisons of thermally contrasting locations and vintages; this approach is of value but is bound to be inconclusive as the effect of temperature is often confounded with other weather and climate factors (solar radiation, vapour pressure deficit and rainfall), management practices and soils. Direct experimental methods comparing fruit grown at a range of temperature are required to prove cause-and-effect, but attempts to modify the thermal regime of the plant often generate secondary effects. Experimental artefacts in controlled environments often include small soil volume, lack of wind and altered radiation regimes, with direct implications for plant physiology and berry composition. Experiments involving controlled temperature in vineyards aim at a higher degree of realism, but are constrained by cost, issues of scale in space and time, and are not necessarily free from artefacts. Indirect and direct methods are of course non-mutually exclusive but complementary. This review critically assesses the methods used to elucidate the effect of temperature on grape berry composition. It emphasises the limitations of studies where confounded effects are overlooked. With the focus on selected berry traits (total soluble solids, total and titratable acidity and anthocyanins), we analyse the dominant effect of temperature and highlight discrepancies and agreements between indirect and direct research methods.