The time spent harvesting chenopod seeds depends both on the method of collection and patch characteristics such as the density and spatial distribution of plants. The simulation assumes that both Chenopodium and Iva were harvested by hand-stripping. There is some archaeological evidence for uprooting of whole plants and storage of Iva as "sheaves" (Jones 1936). In upland situations, chenopod plants tend to occur as small clusters within larger stands dominated by the related C. missouriense (Smith 1987a). Although upland stands may be quite dense, they are small and widely separated. Floodplain stands may also be widely scattered, although density is highly variable. I assume that cultivated plots were planted so as to minimize travel time between plants. To represent harvest times for wild stands, the average rate for ten scattered plants in Smith's collection number 94 (6.7 min/m2, or 1100 hr/ha) was used. Converted values for the same stand if the plants were located at the center of adjacent one-m2 units (4.6 min/m2, or 800 hr/ha) were used for domesticated stands.
Harvest rates for Iva were also based on Smith's (1992) harvesting experiments. The size and density of stands is highly dependent upon soil moisture since Iva can only outcompete other floodplain weeds under relatively hydric conditions. Therefore, it is likely that floodplain populations were more efficient to harvest than upland stands. Domestication has the potential to further reduce travel time between plants. Accordingly, to estimate harvest rates for domesticated sumpweed on the terrace near Skid-more, the highest-density stands harvested by
Smith were used, after discounting samples that had suffered major loss of achenes by the time of collection. Smith obtained a rate of 300 hr/ha for his sample 60, located on an inundated terrace. For wild floodplain stands, an average of six samples was used, eliminating those with substantial achene loss: 0.05 hr/m2 or 500 hr/ha. The density of stands on the hillsides was assumed to be lower due to competition from upland specimens sharing the same habitat. Both cultivated and wild stands in upland settings were likely subject to this constraint, assuming that no culling was carried out. In the absence of a reliable method of more precisely estimating the effects of upland location on density, it was assumed that all upland plots of Iva were harvestable at rates similar to those estimated for floodplain wild stands (500 hr/ha).
Was this article helpful?