|
Fritz Jooste Administrator Posts: 81
2/24/2021
|
Fritz JoosteAdministrator Posts: 81
The manner in which model parameter values are incremented with or without treatments is often a source of confusion. In this help post, we outline a convention that can de-mystify the way in which the JunoViewer model cycles through the model parameter values from year to year. The key concepts to remember when building your model are as follows:
1. Model Parameter Values Represent the condition at the END of the Year
Always consider your parameter values as representing the condition at the END of the modelling year. Apply this concept from the Model Initialization year. The figure below illustrates this concept. In this figure, you can see that the initial (i.e. starting) values provided to the model in the initialization year (year zero) represent the end-of-year condition.
For this first example, we assume no treatments are present. The starting values are 12 years for Age, and 17.5 mm for Rutting. These values are assumed to represent the condition at the end of the model initialization year (year zero). For this example, we assume a simple case where the rut increment is always 0.3 mm per year.
The model now starts in year 1, taking the values at the end of the previous year, and incrementing them. Thus, as shown below, the values at the end of year 1 are 13 for Age (age is incremented by 1) and 17.8 mm for Rut (incremented by 0.3 mm). The same process applies in year 2, etc.
If your models are defined in the DMS, the "Before" Placeholder values at the start of year 1 will hold values of 12 and 17.5. Then, as the model starts applying increments in the absence of treatments, the "After" placeholder values will be updated to 13 and 17.8. Other parameters can be dependent on either the before or after placeholder values.
Note that the model applies increments (and treatments) only from year 1 onwards.
If you are looking the the model output for the above scenario in the Forecast View, and looking at Rut, you will see something like this (presuming that the model initialization year is defined as 2021):
2. Treatments are Assumed to take effect in the Middle of the Year (or somewhere between start and end)
This concept is illustrated in the figure below, which shows a situation in which a rehabilitation is applied in Year 1. We assume the rehabilitation re-sets the age to zero, and the rut depth to 3 mm. After resetting Age and Rut, the model applies normal increments in year 2 since there is no treatment present.
If you are looking at the model output for this segment in Forecast View, and looking at Rut, you will see what is shown below (presuming that the model initialization year is defined as 2021).
Note that the reset values are hidden behind the treatment name. The 3.3 mm is not the reset value, but the value one year after the treatment had been applied.
3. The Model only Starts Applying Increments and Treatment Resets from Year 1 Onwards
You may often have committed treatments in year zero, and then see that no resets are applied for these treatments. This outcome can be understood once you have grasped the concepts discussed above. Since the model only applies treatments and increments from year 1 onwards, any treatments before year 1 are not implicitly taken into account by your model.
There are two solutions to this problem: (1) You can simply set your Model Initialization year back by one year. Note that in this case, we assume the initial values represent the START of year zero (or end of year zero minus one). (2) You can program the necessary logic into your model to adjust the initial values and take into account treatments that occur in or before year zero
|