Generalized Harvest Regime

This behavior can perform harvest very flexibly, depending on the parameters used. The behavior itself decides when harvests will occur and how much to cut, then chooses trees to cut with the help of a preference algorithm.

Trees removed by this behavior will have a mortality reason code of "harvest".

Parameters for this behavior

How it works

The behavior calculates many probabilities to decide whether and what to cut. These calculations can be based on either total adult biomass in the plot (Mg/ha), or total adult basal area in the plot (m2/ha). The parameter What should calculations be based on? indicates which to use. If biomass is chosen, biomass is calculated using the Dimension analysis behavior.

Deciding whether a harvest occurs in a timestep

The behavior begins the timestep by deciding whether or not a harvest will occur. The probability of harvest is calculated as follows:

where:

• P is the probability that the plot will be harvested this time step
• Amt is either the total plot adult biomass (Mg/ha) or the total plot adult basal area (m²/ha)
• a is the Gen Harvest Regime Harvest Probability "A" parameter
• m is the Gen Harvest Regime Harvest Probability "M" parameter
• b is the Gen Harvest Regime Harvest Probability "B" parameter

A random number is compared to this probability to decide whether or not a harvest will occur. The frequency of harvest in previous timesteps is not taken into account.

Deciding the target amount of basal area to remove

Target removal amount is drawn from a probability distribution function. There are two choices: the gamma function, or a user-defined function with ten intensity classes and associated probabilities for each. Choose the function you want with the Distribution controlling harvest amount parameter.

If the gamma is used, the mean of the distribution (as percentage of adult basal area to remove between 0 and 100) is calculated as follows:

where:

• BAR is the mean percent of total plot adult basal area to remove, between 0 and 100
• Amt is either the total plot adult biomass (Mg/ha) or the total plot adult basal area (m²/ha)
• α is the Gen Harvest Regime Remove Amount "Alpha" parameter
• β is the Gen Harvest Regime Remove Amount "Beta" parameter
• μ is the Gen Harvest Regime Remove Amount "Mu" parameter

This mean value is used, along with the Gen Harvest Regime Gamma Scale Parameter, in a draw on the gamma distribution in order to get the actual target removal rate. If a value above 100% is drawn, this value is discarded and the draw is repeated until a value of less than 100% is drawn.

The adult plot biomass or basal area in the equations above is the total for all species. All species must participate in harvest, and only adults are counted and cut.

If the probability distribution function is user-defined, you define ten harvest intensity classes and the probability that the harvest will fall into each of these. Enter the upper bound of each intensity class, as a proportion of total basal area removed, in the User Distribution Intensity Class X Upper Bound parameter. The lower bound of the first class is 0. The upper bound of class 10 is 1. Values must increase with each class but classes do not have to be the same size.

For each intensity class, the value in the parameter User Distribution Intensity Class X Probability is the probability, between 0 and 1, that a harvest within the range of removal values defined by that class will occur. Zeroes are allowed, and harvests within a class of 0 probability will never occur. All ten class probabilities must add up to 1.

This behavior turns the probabilities for all the classes into a cumulative probability function. A random number drawn on the uniform distribution between 0 and 1 is used against the cumulative probabilities to determine which class the harvest falls into. Then a second uniform random draw on the range of values in that class is used to choose the target proportion of adult basal area to remove.

Deciding which individuals to cut

A preference function determines the probability that an individual will be cut, as follows:

where:

• P_i is the probability that individual i will be cut
• α _i is the Gen Harvest Regime Cut Preference "Alpha" parameter for the species of individual i
• β _i is the Gen Harvest Regime Cut Preference "Beta" parameter for the species of individual i
• γ _i is the Gen Harvest Regime Cut Preference "Gamma" parameter for the species of individual i
• μ _i is the Gen Harvest Regime Cut Preference "Mu" parameter for the species of individual i
• BAR is the percent of total adult basal area to remove, between 0 and 100
• DBH is the individual's DBH

The term σ is:

where:

• a is the Gen Harvest Regime Cut Preference "A" parameter
• b is the Gen Harvest Regime Cut Preference "B" parameter
• c is the Gen Harvest Regime Cut Preference "C" parameter
• BAR is the percent of total adult basal area to remove, between 0 and 100

An individual's probability of removal is compared with a random number to determine if that individual is cut.

The preference function takes into account the target basal area removal rate of the plot. However, the function shape does not necessarily produce a mean removal probability equal to that of the target removal rate, particularly near 0 and 100. If you wish, the behavior can refine the probabilities on a second pass to get closer to the target. You can set a tolerance using the Gen Harvest Acceptable Deviation From Cut Target parameter. This is expressed as a proportion of the removal rate, between 0 and 1. So a value of 0.1 allows first-pass acceptable deviation of 10 percent from the target. If the actual removal rate falls outside this limit, the function adjusts all preferences by a correction factor and re-evaluates individual cut decisions. No more than two passes will be made, even if the second pass does not achieve a removal rate within the tolerance. Setting a high tolerance such that a second pass is not often needed eliminates a lot of calculations and will allow a run to be faster.

Saplings can be subjected to an increased mortality rate as a consequence of harvesting. You can turn this on by setting the parameter Do sapling mortality as a result of harvest? to "Yes". (Setting it to "No" means there is no effect of harvest on sapling mortality at all.)

The function for the probability of sapling mortality as a result of a harvest event is:

where:

• PCR is the percent of basal area removed in this harvest, between 0 and 100
• p is the Gen Harvest Regime Sapling Mortality "p" parameter
• m is the Gen Harvest Regime Sapling Mortality "m" parameter
• n is the Gen Harvest Regime Sapling Mortality "n" parameter

Saplings are randomly chosen to die at this rate. All species are affected equally. Saplings are killed with a mortality code of "harvest".

How to apply it

Apply this behavior to the adults of all species. If you choose to make calculations based on biomass, you must also apply the Dimension analysis behavior to the same trees.