DRY-MATTER PRODUCTION IN A TOMATO CROP - MEASUREMENTS AND SIMULATION

Simulation of dry matter production by the explanatory glasshouse crop growth model SUKAM (Gijzen, 1992, Simulation Monographs), based on SUCROS87 (Spitters, Van Keulen and Van Kraalingen, 1989, Simulation and systems management in crop protection), was validated for tomato. In the model, assimilation rates are calculated separately for shaded and sunlit leaf area at different cumulative leaf area in the canopy, laking into account the different interception of direct and diffuse components of light. Daily crop gross assimilation rate (P-gd) is computed by integration of these rates over total crop leaf area and over the day. Leaf photochemical efficiency and potential gross assimilation rate al saturating light depend on temperature and CO2 concentration and are approximated as being identical in the whole canopy. Crop growth results from P-gd minus maintenance respiration rate (R(m); dependent on temperature and crop dry weight), multiplied by the conversion efficiency (carbohydrates to structural dry matter; C-f). Growth experiments (periodic destructive harvests) with different planting dates and plant densities and two datasets from commercially grown crops, were used for model validation. Hourly averages for global radiation outside the glasshouse, glasshouse temperature and CO2 concentration, together with measured leaf area index, dry matter distribution (for calculation of C-f) and organ dry weights (for calculation of R(m)) were the inputs to the model. Dry matter production (both level and dynamic behaviour) was simulated reasonably well for most experiments, but final dry matter production was under-estimated by about 27% for the commercially grown crops. At low irradiance and with large crop dry weight, growth rate was under-estimated, probably as a result of over-estimation of R(m). This could almost completely explain the large under-estimation for the commercially grown crops, which had large dry weight. Final dry matter production was over-estimated by 7-11% if daily averages instead of hourly input of climatic data were used. It is concluded that SUKAM is a reliable model for simulating dry matter production in a tomato crop, except for those situations where R(m) has a large influence on crop growth rate (low irradiance and large crop dry weight). An improved estimate of R(m) would take into account the influence of metabolic activity. A preliminary attempt to relate maintenance costs to relative growth rate (a measure for metabolic activity), showed promising results.