^lADAS Boxworth, Boxworth, Cambridge CB3 8NN, UK; ²Depart. Ciencias do Ambiente, Institute Superior de Agronomia, 1300, Lisbon, Portugal; ³ADAS High Mowthorpe, Duggleby, Malton, N. Yorks., Y017 8BP, UK

Background and objectives
Crop yield is predominantly determined by the crops capacity to intercept light energy and utilize it for growth [1]. Traditionally, empirical models of yield response due to disease have not taken account of host processes nor the wide range of host yield potentials. Thus, the relationship between disease control by fungicides and yield response varies from site-to-site and from season-to-season. The value of disease control is therefore difficult to predict. It has been shown that yield response to disease control can be better explained by taking account of both the amount of green area available for photosynthesis and the amount of incident radiation absorbed by that healthy area [2,3]. However, until recently the practical exploitation of this approach has been limited. Work described in this paper demonstrates how both in-field measurements, and spectroradiometric, remote sensing techniques, could be used to estimate the amount of radiation intercepted by healthy and diseased crop canopies. These measurements could then be used to more effectively explain the variation in yield response to disease control [4].

Results and conclusions
In-field estimates of green leaf area index (GLAI) in healthy and diseased crop canopies were found to be directly proportional to laboratory measured GLAI (R²=0.75). The in-field technique depended on shoot counts and a leaf from factor (F=0.83) which was derived from 20 winter wheat varieties by relating their leaf length and widths to leaf areas measured by planimeter (R=0.95).

In two experiments at ADAS Terrington, on the susceptible winter wheat variety Slepjner, epidemics of yellow rust (Puccinia striiformis) ranged from nil to severe with 60 (1994) and 52 (1995) different combinations of fungicide dose and timing. Assessments of disease severity (%) integrated over time as area under disease progress curve (AUDPC) could only account for yield differences within each season but not between seasons. In-field assessments of GLAI integrated over time, or healthy area duration, showed a curvilinear relationship with grain yield (1994, R²=0.63; 1995, R²=2 0.73). Estimates of intercepted radiation by healthy green tissue (i.e., healthy area absorption HAA) were determined from GLAI and total incident radiation measurements using the Beer's Law analogy (k=0.5). Accumulated intercepted radiation by green tissue accounted for more variation in yield response to disease control with R²=R²0.80 (1 994) and R²=0.92 (1995).

Spectral reflectance measurements of both healthy and diseased (P. ;striiformis) crop canopies were made using a LICOR LI-1800 spectroradiometer over the wavelength range 350-800nm at 2-nm intervals. Contrasting spectral reflectance signatures were obtained for healthy and diseased crop canopies. Differences in spectral reflectance signatures within the visible spectrum related to the amount of incident radiation absorbed and utilized by the crop canopy. Changes in the infrared spectrum related to changes in leaf water content owing to the presence of disease. It is suggested that measurements of spectral reflectance, using remote sensing techniques, may be exploited in the future to relate the absorption of radiation by green tissue over time to yield response due to disease control.

References
1. Monteith JL, Unsworth MH, 1990. Principles of Environmental Physics. London: Edward Arnold.
2. Waggoner PE, Berger RD, 1987. Phytopathology 77, 393-398.
3. Bryson RJ, Paveley ND, Clark WS, Sylvester-Bradley R, Scott RK, 1997. European Journal of Agronomy 7, 53-62.
4. Hansen JG, 1991 EPPO Bulletin 21, 651-658.