Productivity of Agave deserti: measurement by dry weight and monthly prediction using physiological responses to environmental parameters

Summary An environmental productivity index based on physiological responses to three environmental variables was used to predict the net productivity of a common succulent perennial of the Sonoran Desert, Agave deserti, on a monthly basis. Productivity was also independently measured in the field from dry weight changes. The index was based on soil water availability, day/night air temperatures, and photosynthetically active radiation (PAR), which were individually varied in the laboratory and the effect on net CO₂ uptake by the leaves determined. From monthly precipitation, temperature, and PAR at the field site together with the responses measured in the laboratory, an index (maximum value of unity) was assigned to each of these three environmental variables and their product was termed the environmental productivity index. This index indicates the fraction of maximal CO₂ uptake expected in the field for each month (well-watered A. deserti assimilated 285 mmol CO₂ m^-2 leaf area day^-1 at PAR saturation and optimal day/night temperatures of 25° C/15° C). The dry weight analysis was based on the monthly unfolding of new leaves from the central spike of the rosette and their seasonal increase in dry weight, which were determined in the field. The production of new leaves was highly correlated with the environmental productivity index (r²=0.93), which in turn was highly correlated with the water status index (r²=0.97). After correction for respiration by folded leaves, stem, and roots, plant productivity predicted by the average environmental productivity index (0.36) over a wet June-to-October period agreed within 4% with the productivity based on the conventional dry weight analysis. The net productivity of A. deserti over this 5-month period was 0.57 kg m^-2 ground area (5.7 Mg ha^-1), a large value for a desert CAM plant. The environmental productivity index proposed here may provide a reliable means for predicting net productivity on a monthly basis, which may be particularly useful for species in relatively variable environments such as deserts.