| |
DEFINITION AND ESTIMATION OF TOP HEIGHT FOR SITE INDEX
4.0 Recommended Top Height Definition
TAC recommends that Option 1 be used as the top height definition. We restrict the use
of top height for defining height and age for site index. It is recognized that the height
of other trees may be required for other uses.
Although the definition of top height is arbitrary, Option 1 is the most robust among
the alternatives. It is simplest in concept and hence least ambiguous to follow. It is
also the cheapest to implement in the field. In short, it is the definition most likely to
be measured and applied in an unbiased and consistent fashion. Option 1 is similar to
Option 2 but does not allow smaller diameter trees to be selected and is restricted to
valid site index trees.
Regarding the issue of plot size, the elite cohort of trees we seek to measure and call
top height trees is dependent on plot size. Estimates of top height change with
changing plot sizes and the degree of change is difficult to quantify. In addition, as
plot size increases so does the cost, the chance of choosing incorrect trees, and the
number of invalid trees (e.g., largest trees but with broken tops). The invalid trees are
not always randomly distributed among the "elite cohort" providing a further
source of bias.
The 0.01 ha plot size is modular enough to be practically used in young stands (as part
of silvicultural surveys) as well as in older stands allowing all field sampling to
estimate top height, age and site index consistently. The 0.04 ha plot is totally
impractical in young stands and likely not practical as part of an operational inventory.
The costs at both levels are exorbitant and the fieldwork entailed would be extremely
grueling.
The intent to measure more trees on a bigger plot has resulted in a definition that
cannot be consistently measured or applied (other than in theory). From a data legacy
viewpoint, TAC concludes that, with the exception of the VRI definition, top height has
never been measured in an unbiased fashion in B.C., making post facto correction
all but impossible.
The reason for calling the plot a "null" plot if the largest diameter tree is
not valid is that any smaller trees do not represent the site potential. Similarly, the
reason for selecting the largest diameter tree, regardless of species, is that it is the
only consistently identifiable tree that represents the site potential. Moving down in
size to obtain a suitable tree based on a preferred species will result in biased
estimates of top height and site productivity. Trees other than the largest diameter tree
can (and probably should) be measured, but they should not be called top height trees.
Because our new definition must be applied province-wide and for a variety of
conditions, we chose to base it on the largest diameter tree, although the tallest tree is
perhaps the ideal theoretical largest tree. The primary reason for choosing the top height
tree based on diameter rather than height is that the largest diameter tree can always be
identified with certainty on a given plot.
- There are two important (and self-evident) stipulations necessary to ensure the
definition is applied in an unbiased fashion:
- If a valid tree (i.e., a tree that can be used to estimate site index) is not present in
the plot, then top height is not measured (a 'null' plot); and
- The largest dbh tree is eligible as a top height tree, regardless of species.
- Inventory concerns regarding the need for data on leading species and species in
different layers can be simply addressed by measuring extra trees in the plot (but these
are not top height trees using Option 1) and establishing additional plots if necessary.
Specifically, we differ slightly in our definition from the VRI definition in not choosing
smaller trees if the largest is invalid and by excluding dead or down trees, which are, by
definition, not appropriate site index trees. Any additional trees measured for height in
VRI plots should not be called top height trees.
- For a majority of situations, the largest diameter tree is the only tree that can be
identified with certainty ensuring a measurement consistent with definition and
application. In some situations, such as some young stands on flat to moderate slopes, TAC
recognizes that the tallest tree may be easier to identify. However, although measurement
could be internally consistent under these conditions, the estimate is a biased estimate
of Option 1. In some mixed species stands, the largest diameter tree may not be part of
the upper canopy layer, in which case, it would be an invalid site index tree.
- Field procedures across all business applications will need to be standardized.
- The current requirement for a "top height" for green-up requirements (BC Reg
107/98) may or may not be consistent with the definition given in Option 1. Every
stand is now legally required to have a "top height" for green-up. This may be
inconsistent with TAC’s proposed definition if the tree is not an appropriate top
height tree, i.e., the leading species does not have a measurable top height or the trees
have had their height growth affected by repression.
- There will be a one-time, higher cost of re-measuring PSPs as they will need to be
sectored and top height trees identified. It is recommended that all trees previously
measured for height continue to be measured. Top height trees should be added as
additional height trees if they are not already being measured.
- The implications surrounding the use of current models are as follows:
- TASS uses predominant height and adjusts top height by an approximation. TASS can be
made compatible with the proposed definition. (Because field measurement of top height in
permanent plots has been done inconsistently in the past, there is a danger that model
estimates are also biased and inconsistent. This diminishes as top height is measured
correctly).
- VDYP is a legacy model incompatible with the current definition; its replacement,
Inventory Projection System (IPS), can be made consistent with the new definition.
- The impact of changing the top height definition on existing site index curves has not
been examined as part of this analysis. As a starting point, TAC suggests that the general
height-age trends for extant models will not change markedly over much of the range of the
data. This implies that current height-age curves can continue to be used unless they can
be shown to be inappropriate.
- TAC recognizes that there will be discrepancies between older legacy data (e.g., VDYP,
temporary plots) and newer endeavors (e.g., VRI) and that this should not limit the
adoption of the proposed definition. It is scientifically defensible to base models on
consistently measured, unbiased drivers such as top height, rather than force field data
collection to be based on the vagaries of previous sample plot data collection
methodologies.
Previous page | Next page
Last updated: 01/13/99 |