| Abstrakt: |
Freezing injury can cause extensive yield loss in alfalfa (Medicago sativaL.) exposed to severe winters in cold climates. Alfalfa models must simulate freezing injury effects in a cumulative way, so that forage dry matter (DM) yield can be accurately calculated during multiple years of a crop. A companion paper presents ALFACOLD, an alfalfa simulation model developed by integrating functions of cold tolerance, fall dormancy, and freezing injury into the ALSIM 1 (Level 2) alfalfa model. The ALFACOLD model calculates forage yield on a daily basis, while simulating freezing injury across multiple years of the same crop. The objective of this study was to test the ALFACOLD model's ability to calculate field‐measured forage DM yield. The model was tested against measured forage yield from 39 cultivars seeded in two years and managed under a 4‐cut system for 3 yr after the seeding year at two sites in Wisconsin. Averaged across the sites, cultivars, and years, the ALFACOLD model calculated annual forage DM yield to within 12% of the measured yield, compared with an error of 35% with the ALSIM model. A regression of field‐measured yield of individual harvests (n= 874) on the corresponding ALFACOLD calculated yield indicated that 70% of the measured variability in forage yield was explained by the model. The average difference between measured and calculated DM yield for Harvests 1, 2, and 3 was <7%, but the model overcalculated fourth‐cut yield by 38%. Averaged across all sites, cultivars, years, and cuts, the ALFACOLD model calculated yield with an average SD of 600 kg ha−1cut−1. |
Plný text