1) Trees in thinned areas had much higher levels of foliage after accounting for variability in tree size (dbh) and crown ratio. Thus the allometric relationship for foliage biomass in ponderosa pine is not stationary; the relationship changes with respect to thinning treatment. The dependence on treatment for foliage biomass implies that reliance on stationary allometric equations may result in substantial error when estimating crown biomass in ponderosa pine stands.
2) It appears that site productivity may have a larger effect on foliage mass per tree than we first anticipated. Although the evidence is somewhat anecdotal (all from one Experimental Forest), trees on a slightly higher site (the Gravel thinning T8) maintained much higher levels of foliage per tree. A closer examination of the Blacks Mountain Ecological Unit Inventory indicated that this area of the forest represented slightly higher levels of productivity.
3) Eight and ten years after thinning, annual foliage production is approximately 27 percent of the total. Since stands appear stable at this point, this also may be considered an approximate level of needle cast after stands have rebuilt crown.
The higher rate per unit of foliage two years following treatment is to be expected since during that period, the trees are rebuilding crown and increasing the amount of foliage in retained trees. By years 8 and 10 the trees appear to have fully responded and foliage production per unit of biomass has decreased. However rates per tree have actually increased as total foliage for a tree of given dimensions has increased in response to thinning.
4) Foliage is maintained in unthinned areas for about 4 years usually with needle retention extended for a year or two post thinning in the base of the crown. This extension of needle retention implies a short-term drop in litterfall in the years immediately following thinning resulting from both the overall reduction in crown mass in the stand combined with the increased retention.
5) In a referenced fuel reduction thinning at Blacks Mountain Experimental Forest, these equations indicate that five years after thinning, foliage biomass at the stand level has already recovered to pre-treatment levels. Thus the crown foliage removed from the understory appears to simply have been shifted up into the crown, and this process happens very quickly.
This stand was thinned from below. Before treatment, the stand had a basal area of 16.5 m2ha-1, (s.e. = 1.02 m2ha-1) and the quadratic mean diameter was 10.3 cm. After thinning from below, the number of stems was approximately 112 ha-1 and basal area was reduced to 10.6 (s.e. = 0.89). This thinning increased quadratic mean diameter to 35 cm. Five years post-thinning, basal area was 12.1 (s.e.=1.01), well below the pre-treatment level yet, the foliage biomass has recovered completely.
Thus after thinning, total foliage biomass and the amount of foliage created (and produced as litter) at the stand-level recovers quickly. The long term fire-hazard benefit to this thinning comes from the re-arrangement of fuels: the reduction in ladder fuels. Litterfall rates, even in a heavily thinned stand such as this where stems per ha was reduced from 2000 to 112, are still going to produce needle cast for surface fuels at or near pre-treatment levels over the long term and these widely spaced stands may still carry frequent surface fires.
In practice, biomass estimation often relies upon extrapolated equations. The working assumption being that a single allometric equation for a given species should be sufficient, or at least any errors due to extrapolation are irrelevant. Unfortunately, the allometric biomass relationships are not likely to be stationary over space, or even over time within a given locale. This analysis suggests that changes in stand structure due to thinning will change the allometric relationship in ponderosa pine. We also found some evidence that for foliage, site productivity may be an even greater concern. If we were to mistakenly use the foliage equation from the T8 (Gravel thinning) in an unthinned stand on the west side of Blacks Mountain Experimental Forest, we could obtain foliage biomass estimates that were off by nearly 100 percent. If there is that much variability within a fairly uniform 4,000 ha forest, imagine what errors might be created by applying these same equations in, say, the west side of the Sierra Nevada.
For ponderosa pine, at least, treatment regime and site productivity should be considered when developing estimates of biomass, particularly crown biomass. And equations should only be applied for similar sites and treatments. Branch and bole biomass appear to be much more stable and equations may be more reliably extrapolated for these components of biomass.