General circulation models (GCMs) project increases of the earth's surface air temperatures and other climate changes in the middle or latter part of the 21st century, and therefore crops such as cotton (Gossypium hirsutum L.) will be grown in a much different environment than today. To understand the implications of climate change on cotton production in the Mississippi Delta, 30 years (1964 to1993) of cotton growth and yield at Stoneville, Mississippi, USA, were simulated using the cotton simulation model GOSSYM. The GCM projections showed a nearly 4°C rise in average temperature and a decrease in precipitation during the crop growing season. The fertilization effect of an increase in atmospheric CO 2 concentrations from 360 to 540 ppm, without the change in other climatic variables, increased yields by 10% from 1563 to 1713 kg ha -1 , but when all projected climatic changes were included, yields decreased by 9% from 1563 to 1429 kg ha -1 . The rate of plant growth and development was higher in the future because of enhanced metabolic rates at higher temperatures combined with increased carbon availability. The effect of climate change on cotton production was more drastic in a hot and dry year. Since most of the days with average temperatures above 32°C will likely occur during the reproductive phase, irrigation will be needed to satisfy the high water demand, and this reduces boll abscission by lowering canopy temperatures. Therefore, if global warming occurs as projected, fiber production in the future environment will be reduced, and breeding heat-cold-tolerant cultivars will be necessary to sustain cotton production in the US mid-South. Cultural practices such as earlier planting may be used to avoid the flowering of cotton in the high temperatures that occur during mid to late summer.

The mechanisms describing leaf appearance and tillering are vital to the modeling of wheat canopy development. How these two factors will be affected by increasing global atmospheric [CO2] in cool or warm climates is not fully understood. Two southeastern USA adapted wheat (Triticum aestivum, L.) cultivars, Coker 762 and Stacy, were grown under nearly nonlimiting conditions including elevated [CO2] (600 μL L−1) and under six air temperature regimes (ranging from 4/−1 to 18/7 °C d/night and progressively increasing to 16/4 to 29/18 °C d/night during the season) to observe leaf and tiller appearance rates and to compare tillering rates to those predicted by the Fibonacci series as approximated by Binet's equation. Both cultivars exhibited an abrupt one‐time change in their phyllochron interval for all six temperatures. This change occurred just prior to double ridge formation. The vegetative growth phase phyllochron interval of the two cultivars was significantly different only in the (21/10 °C) temperature treatment. In the two lowest temperature treatments (16/4 and 18/7 °C), the cultivars differed in phyllochron interval during the reproductive growth phase. The tillering rate of wheat followed closely the theoretical development predicted by Binet's equation during the vegetative phase of development.

No abstract

The empirical soil temperature subroutine of the cotton simulation model GOSSYM did not perform well under bare and cotton‐cropped (Gossypium hirsutum L.) surface conditions in the field. Therefore, it was replaced by a mechanistic soil temperature subroutine called HEAT. The HEAT subroutine was validated against soil temperature data collected in the field under bare and cotton‐cropped surface conditions, and was compared with the empirical TMPSOL subroutine of GOSSYM. Soil temperatures were collected in the field under bare and cotton‐cropped surface conditions at two locations (row and furrow) and four depths (5, 10, 25, and 60 cm) during the growing seasons of 1991 and 1992. Under bare surface conditions, HEAT underpredicted the average daily soil temperatures by 1 to 7°C at all locations and depths. However, under cotton‐cropped surface conditions, HEAT calculated the soil temperature adequately, especially after canopy closure. Under bare surface conditions, the temperatures calculated by TMPSOL were closer to the measured values than those calculated by HEAT, but were still beyond the standard deviations of the measured values. It is recommended that the TMPSOL subroutine in GOSSYM be replaced by the HEAT subroutine.

No abstract