Optimizing thermal imaging as a technique for detecting stomatal closure induced by drought stress under greenhouse conditions Olga M. Grant a *, M. Manuela Chaves a,b and Hamlyn G. Jones c a Laborato ´rio de Ecofisiologia Molecular, Instituto de Tecnologia Quı´mica e Biolo ´ gica, Apartado 127, 2781-901 Oeiras, Portugal b Departamento Bota ˆ nica e Engenharia Biolo ´ gica, Instituto Superior de Agronomia, Universidade Te ´ cnica de Lisboa, Tapada da Ajuda, 1349-017 Lisboa, Portugal c Plant Research Unit, Division of Environmental and Applied Biology, School of Life Sciences, University of Dundee at SCRI, Invergowrie, Dundee DD2 5DA, Scotland, UK Correspondence *Corresponding author, e-mail: olga.grant@emr.ac.uk Received 23 June 2005; revised 2 December 2005 doi: 10.1111/j.1399-3054.2006.00686.x Temperature of leaves or canopies can be used as an indicator of stomatal aperture, which is considered a sensitive response to soil water deficit. In this paper we analyse the robustness and sensitivity of thermal imaging for detecting changes in stomatal conductance and leaf water status in a range of plant species. Thermal imaging successfully distinguished between irri- gated and non-irrigated plants of a variety of species under greenhouse or controlled chamber conditions, with strong correlations between thermal indices and stomatal conductance measured by porometry. Our results also highlighted issues that need to be addressed in order to be confident of always detecting drought stress using this technique. These include variabil- ity in leaf angles and the limited reliability of ‘wet’ and ‘dry’ leaves to represent leaves with stomata fully open or stomata fully closed. These results should assist the design of protocols for application in crop produc- tion or ecosystem monitoring. Introduction The early diagnosis of stress in plants has been hampered in the past by the need for destructive sampling or intensive in situ measurements. Imaging approaches using reflectance, fluorescence, or thermal sensors have been applied since the early 1980s (e.g. Omasa et al. 1983, Nilsson 1991, Chaerle and Van Der Straeten 2000, Chaerle et al. 2004, Riera et al. 2005) for non-destructive and non- invasive monitoring of the physiological condition of plants. Use of these imaging technologies in the field (e.g. Jones et al. 2002) can help increase efficiency in crop manage- ment and crop breeding (Chaves et al. 2003) or assist the monitoring and management of natural ecosystems. However, these new imaging techniques must be validated and calibrated against classical physiological analysis of gas exchange and water relations (Chaves et al. 2003). Under limited water availability, reductions in stoma- tal conductance can occur even before any change in plant water status (Davies et al. 2000), meaning that monitoring stomatal conductance can be a better indi- cator of plant responses to drying soil than monitoring water potential (Jones 2004b). However, cuvette methods for measuring stomatal conductance are time-consuming, labour-intensive and only give point measurements. Thermography, based on the principle that as water is lost through the stomata, leaves cool, and hence as stomata close, leaf temperatures rise, is a promising alternative. The objective of research in this area must now be to define the limitations and refine the technique so that its Abbreviations – ANOVA, analysis of variance; CWSI, crop water stress index; g s , stomatal conductance to water; T, temperature. Physiol. Plant. 127, 2006 507 Physiologia Plantarum 127: 507–518. 2006 Copyright ß Physiologia Plantarum 2006, ISSN 0031-9317