The survival probability for trees is predicted and then multiplied by the expansion factor to determine the surviving number of trees per ha.
The basic estimate of annual probability of survival is provided by a logistic function based on the work of Yao (1996), Yang (2001) and recently modified (2006). Major predictive variables: DBH, DI, BA, BA in larger trees (total and by species groups), species composition (by basal area). For juvenile tree mortality see here.
Ideally a well-structured logistic function should capture mortality dynamics, such as recreating self-thinning and old-growth behaviour. However, since the western boreal has some significant data gaps, additional constraints are required to ensure realistic mortality rates. Because of this, MGM accelerates mortality when:
- Trees are above a maximum size-density limit (the self-thinning line). Stands with unusually large initial stand densities should experience heavy mortality to bring them back below this empirical line. The maximum limit is for all species on all sites and was not separable by site classes based on Alberta PSP data.
- Deciduous site trees that reach the upper end of the site index curve have their mortality rate increased by 10% to simulate old stand break-up.
- Stand basal area/ha is above a maximum stand basal area for the species. These maximum basal areas were determined based on PSP data using the upper basal area percentiles. The maximum basal area is different for different species in pure species stands. For mixedwood stands, there is one common maximum basal area.
Constraints 2 and 3 place yield limits on old deciduous stands to simulate break-up. Recent literature (Kabzems and Garcia 2004 Can J For Res 34:384-395) suggests that the Fort Nelson region of NE BC does not experience this breakup due to a moister and cooler climate. Users in this region may wish to switch constraints 2 and 3 off by removing the checkmark from the "MA Flag (Stand breakup adjustments)" under the MGM Crop Plan Options Event - Other Settings tab.