|Summary||On forest carbon cycle||
How to run
|General description||Examples for model output||
HOW TO RUN CASMOFOR?
2. In case the rate of afforestation by total area, species or yield class changes over time, fill in the appropriate table of the user.xls file, and save the file. (Click here to get help on this.)
3. Click on CASMOFOR.xls to run the program.
4. Select your language.
5. Select the way the program is to be run:
6. For the option selected above, provide input information as requested in two consecutive input windows.
7. If any of the species you want to model cannot be found in the list offered in the pervious step, select one with possibly similar growth and other characteristics. If there is no such a species, you need to add all data in the knowledge base of the model. Contact the developer of CASMOFOR, Zoltan Somogyi, to do this. After this has been done, go back to Step 3 above.
8. Wait until all calculations are done.
9. If option a) above has been selected, you can directly check the results by switching to many graphs, or you can check the knowledge base of CASMOFOR.
10. After the program terminates for all options above, the file containing all results (named "results_*.xls", where you must provide a name for *) remains open. In this file, you can check all inputs you provided, as well as all results either in tabular, or graphical format, by selecting the various sheets whose list can be found at the very first sheet of the file. See also the "Managing outputs" secion.
Please make sure that you have read also the instructions on installation.
In the section below, it is demonstrated how CASMOFOR can be used to address issues that are developed in the "Developing scenarios" section.
Note also the section to demonstrate what questions can be easily answered by running CASMOFOR by including some examples in the Analysing results section.
was developed to enable one to estimate the amount of carbon that is
sequestered, and emitted, by a given set of forest stands over time. After
specifying these stands and some other parameters, the model calculates, in
about half a minute - two minutes, the carbon stock changes for the specified
period of time, and produces MS Excel graphs and tables that the user can use
just like any other MS Excel graphs and tables. The calculated carbon removals
and emissions are given for the total forest area by species or carbon pools
(above-ground biomass, below-ground biomass, deadwood, dead roots, litter, soil,
product pools, or the five IPCC pools as defined
in Table 1, on page 1.9, of Volume 1, Chapter 1 of the IPCC 2006 Guidelines).
CASMOFOR is a set of worksheets and graphs also including many accounting functions that are based on the results of many years of forest research, assumptions and expert judgment. These functions are interlinked, and are represented in the worksheets by MS Excel formulas.
In order that the functions are correctly used, the model is parametrized for the Hungarian conditions. This means that all quantified knowledge is included in the model that are available for the main Hungarian tree species and that are relevant for the carbon cycle of the forest. In addition, the user must define the afforestation program to be analysed. An afforestation program of one tree species, which may be part of a more complex program, is called a scenario. An afforestation project may include one or several scenarios. Before conducting the simulation of forest processes, these scenarios must be defined. After this, the program makes all calculations necessary, and produces a set of MS Excel sheets and graphs that represent all relevant results. These result can be further analysed by the user according to his or her needs.
A scenario is a set of conditions in the modeled forest that is assumed to prevail during a given period for which the model is run. As no one knows exactly what really happens in forests (first of all if we want to predict what happens in future), we can only work out some generalized and simplifying assumptions, based on which we can define those conditions that considerably affect the performance of the system under analysis. In case of CASMOFOR, this system is how the carbon stocks of an afforested land change, and scenarios are sets of conditions that affect tree growth and other processes that result in carbon stock changes.
In order to build scenarios, one first has to know all major factors that may affect carbon stocks. Not necessarily in order of importance, those factors that can be changed in CASMOFOR, in one way or another, are listed below. For practical reasons, these factors are listed in three groups. The first group includes factors that indirectly affect carbon stock changes. These are the following:
These factors affect carbon stock changes, as well as related forestry characteristics such as costs and revenues, by influencing other factors that in turn directly affect forest processes.
Those factors, listed in the second group below, are those that depend on the above three factors, i.e., that may be species, site and/or age dependent, and that have direct effects. They are the following:
Also species dependent are the following economic information:
Finally, there are factors that do not depend on any of the above factors (at least as modelled in CASMOFOR):
Under certain conditions, some of the above factors could be changed. Here, a clear distinction should be made between setting the parameters of the model (also called model calibration), which cannot be changed by the user, and which are inherent characteristics of the forestry systems, and defining relevant characteristics of afforestation projects that are the implementations of various land use policies. Within an afforestation project, several species-specific scenarios can take place that are defined by selecting appropriate factors that affect carbon cycle.
Note that, in a scenario in the sense of modelling by CASMOFOR, you can only have one species. However, for practical reasons, scenarios of single species are grouped into batches (see below) so that a set of maximum three scenarios can be run at the same time to save time. An "afforestation scenario" can consist of many species, and simulating such an afforestation scenario requires running batches of one to three species each:
An example for setting the model parameters is to apply the growth characteristics of a species, based on forest yield studies. These characteristics cannot be changed in the afforestation project, rather, they are inhrerent parameters of the model. The aim of setting these model parameters is to best model carbon stock changes of the above species in a specific afforestation scenario. CASMOFOR includes parameters that are thought to be applicable in typical Hungarian conditions. In order that the model can be applied in any specific case, Hungarian or otherwise, where local parameters are known, any of these model parameters can be changed. Moreover, parameters of species can also be incorporated in the model that are not included in CASMOFOR, thus making it applicable in any country. However, such a parameter setting should not be viewed as developing scenarios, rather, as calibrating the model.
In contrast, when conducting an afforestation project, we can define several characteristics of the afforested system that are external to the model, and/or that can be changed in the project. For example, we can set the area to be afforested, and the species to be afforested, or the afforestation rate of a certain species in an afforestation program (characteristics that are external to the model). We can also change certain characteristics of the thinning regime or the age of final cutting, which are part of the model, but can depend on the intention of how the afforested system is to be managed. When running the model, we can define any of these characteristics, which is called developing scenarios.
Scenarios can simply be built by setting and changing those conditions of a possible afforestation project that are usually set during the planning and the implementation phase of such projects. Scenarios must be developed by forming simple questions that one may want to answer by running the model of afforestation projects, i.e. CASMOFOR. The most important possible types of scenarios are listed below. When running CASMOFOR, input values associated with these questions can be set by the user to address any of these issues separately, or combined. It is also to be noted that the answers can highly depend on the time frame of the analysis, i.e. for how long the processes are analysed into the future. Therefore, the width of the projection window, or projection length, can also be set.
For examples of results, which are produced by CASMOFOR and which can be used
to answer the questions, click on the "Example result" after each question.
The first form of
the question is simply answered by setting different (total) sizes of project area. The area can
change e.g. from a couple of dozens to several thousands, depending on the
land that is available for afforestation. However, in theory,
CASMOFOR can model
afforestations of millions of ha. -
This question is automatically answered by CASMOFOR after running the program by using any scenarios. Depending on the way CASMOFOR has been run, either all compartments are shown on a single graph, or a separate graph is produced for each compartment. Note that aggregate data of the IPCC pools are also calculated. - Example result
Afforestation campaigns are usually done in several years,
not just in one year. The amount of sequestration depends on the temporal
distribution of afforestation, i.e. whether more areas are afforested in the
first few years and less afforestations are done later, or vice versa.
CASMOFOR allows many
different schemes to handle how much area is afforested and when. The simplest
scenarios is when the same amount of area is afforested each year for a set
number of years. For more
grow at a different overall rate, and at different rate over time. For example,
black locust and poplars are fast growing species, whereas oak and beech are
slow growing ones. By selecting species one can affect these growth rates.
It must, however, be noted that species selection is a rather difficult task, as
the occurrence of species is site specific, so not all species can dwell on all
sites. For more details, see below. -
above, tree growth rate is site-dependent. However, almost all processes in
forests are site dependent as they usually depend on temperature, water supply,
nutrient availability and their distribution over time and space. However, due
to lack of data, site dependency in
CASMOFOR is manly limited
to site-dependent growth rates and silvicultural regimes. You find more details
on site-dependent growth rates
here, and on site
in general and
here. The effect of site is
simply modeled by specifying how much of the various species are afforested in
the various yield classes.
trees, one usually has to do soil preparation. This is an operation during which
soil is disturbed, and it may loose some carbon. To see the effect of this loss,
one may set the amount of carbon lost to some value. For more details, click
or here. -
Carbon from soil
can also be lost if grasslands are converted to forests. In contrast, converting
cropland to forest usually does not involve loosing soil carbon. This has been
demonstrated by Horvath (2006) and Somogyi and Horvath (2006). To analyse the
effect of this phenomenon, set the ratio of grasslands to croplands. Note
that, in the last decades, about (75-)90% of all afforestations in Hungary were
made by converting cropland to forest land. For more details, see in NIR Hungary
It is sometimes assumed that the carbon that is sequestered for some time in the forest, is easily lost due to disturbances, and even the conclusion is sometimes made that, because of the above, it is not worth doing afforestations. This question can be analysed by setting a long enough projection period, during which the long-term dynamics of the amount of carbon sequestered can be seen. The projection period must be longer than the length of the afforestation campaign plus the lenght of the rotation period of the longest living species in the batch. - Example result
Mortality is natural dying of trees due to disturbances such as wind, snow or ice break, fire or others. Such disturbances can occur even in managed forests. By assigning certain probability values based on past experience, one can analyse to what extent such disturbances may have an effect on the carbon sequestration capacity of the forest.
- Example result
Cost is an important aspect of taking
decisions. Therefore, it became imperative to make it possible to analyse both
the magnitude of costs, as well as when they occur. The database of the model
includes net cost information on all major forestry operations. This information
is species and site dependent. -
CASMOFOR, as all models, is subject to errors and mistakes. Mistakes were tried to be eliminated, or minimized, by testing the model under various conditions. There are two types of errors: systematic and random ones. Systematic errors are those that take the same values any time a model parameter is used in the calculations, whereas random errors affect the parameter values on a random basis. CASMOFOR can be used to check the possible effect of random errors on the results of the simulation by randomly changing the values of selected parameters. These effects can be analysed by running any one batch, defined above, many times. - Example result
model parameters are stored in MS
Excel files. These parameters are inevitably average values, which cannot always be the "best"
in specific cases. CASMOFOR
includes parameters for the Hungarian conditions. When a specific case must be modelled,
and case-specific parameters are available, these can be used instead of the
default model parameters, and can be inserted in the model. In order to do that,
load "user.xls" file. You can set each parameter of the model. Growth, thinning
and mortality data can be set in one sheet, and all other parameters in another
one. You can only set parameters so that you include all data for a species. If
you do not have a data for a parameter, you can use the default data or an
expert judgment. You can use the Monte Carlo analysis (sensitivity analyisis)
modul of the model to see the potential effect of the uncertainty of the
individual parameters on the uncertainty of the carbon stock change estimates.
The model can be run in several ways from setting only one scenario, when one is interested in the amount of carbon sequestered, to running several scenarios and adding up/comparing the results, i.e. the effects of the various scenario settings. However, the model can also be run by setting one scenario and changing some model parameters many times in a random way to analyse the effects of possible model parameter errors (Monte Carlo analysis, see below).
Note that, in order to minimize computer time, CASMOFOR was originally designed to run, and then compare or add up up to three scenarios in one run. A set of these scenarios is called a batch. For each batch, there are settings that are species independent, and that thus are the same for all scenarios, e.g. the projection length, however, by selecting up to three species in one run, up to three species-dependent scenarios can be defined: by setting the species, one must select the yield class distribution, and species-dependent parameters are also set automatically in the model by selecting the species.
Running one batch of up to three scenarios. Select this option if you want to run up to three afforestation scenarios at a time in one batch with model parameters that have been set before running the model. The objective of this analysis is to estimate emissions and removals for the total afforestation area over time by up to three species, as well as by all pools separately, and for the total of all three species.
Running several batches of scenarios and adding them up or comparing them. Select this option if you want to add up or compare more than three scenarios, based on the estimated emissions and removals, by using the same (pre-set) model parameters. The batches can typically differ either by area, species or yield class distribution, but also by any other project characteristics that can be changed such as the silvicultural regime, or how the afforestation affects the soil of the afforested area. For each batch, you can set different species independent values (e.g. project area), and up to three species dependent settings. In one run, up to ten batches can be set. Note that it is the scenarios, not the batches, that are compared. If you want to compare the results of the totals of several batches (i.e. if two afforestation projects of several scenarios are to be compared), then first you need to obtain the totals for a group of batches in one run, which produces one results file, then you similarly need to obtain the totals for another group of batches in another run, which produces another results file, etc., and the you must compare the data and/or graphs of these runs yourself.
Running one batch of scenarios many times with different model parameters as defined in a Monte Carlo analysis. Select this option if you want to analyse the effect of random errors in the parameters on the estimated emissions and removals for up to three scenarios. This analysis can also be called a Monte Carlo analysis. In this analysis, the value of certain parameters are randomly modified in each running of the same scenario(s). The parameters to be changed are selected by the user. During running of the model, random numbers are generated to pick values from the error distribution of the selected parameters. This distribution are normally distributed around the pre-set (default, or user-set) mean of the distribution, which is the value of the parameter as set in the model. See also Chapter 220.127.116.11 of Volume 1 the IPCC 2006Guidelines.
The program is run by clicking on CASMOFOR.xls in the file manager of the Windows operating system, or by opening this file from MS Excel.
Note that, as it was mentioned above, the user may want to change some model parameters, or add parameters to species not included in the model. In order to do thi, contact the developer of CASMOFOR, Zoltan Somogyi. If any parameters are to be changed, then this must be done before running CASMOFOR.xls. After CASMOFOR.xls has been opened, no model parameter can be changed.
When the opening window appears, the language by which the model communicates must be selected (English or Hungarian).
Then, a welcoming screen appears from which various help can be accessed by clicking on various buttons, or the next step can be invoked.
The next window is used to select the type of running, i.e. running one batch, or several of them and adding them up or comparing them, or running a sensitivity analysis (see above).
This sheet allows users to provide necessary input to the Monte Carlo analysis, and to an analysis of the effects of a possible bias in a model parameter. The input for the Monte Carlo analysis is the assumed standard deviation (in %) of a model parameter. For each parameter, three input boxes are provided for the three tree species. For the Monte Carlo analysis, the fourth (green) box should contain zero (0). If this fourth box contains 1, it means the figures in the first three boxes will be used as bias. For example, if you suggest that your tree growth is by 20% higher than that of the yield tables, you must enter the number 20 in the first three columns (as appropriate) and the number 1 in the fourth box, however, if you just assume a possible random error in the growth rate data, then enter the 20 in the first three columns (as appropriate), and either put 0 in the fourth box or leave it empty.
Note that the maximum value for the first three columns is 30. This is because if the standard deviation of an error distribution is 30 it means that individual deviations from the mean (which are produced by using the random number generator of MS Excel) can very easily reach 100%, which may cause the program to crush or produce unrealistic output. In order to prevent this, the program limits the individual deviations below 100%, which may cause some deviations to be less "random", the resulting error distribution to be a bit less "normal" than the "normal distribution", and the actual standard deviation of the generated errors, however, this may have a minor or no effect at all on the outcome of the sensitivity analysis.
The next screen is the first part of defining general scenario and afforestation characteristics where you provide input for the simulation:
Note that you can continue from this screen using one of two options. In most cases, simply continuing is the required option. In some cases, specific settings can be made. These settings currently include selecting natural disturbances instead of harvesting in all thinnings, and sensitivity analysis of the soil carbon functions (for specific details, see the input screen).
After this screen, the simulation of the batch is perfomed. In case several batches are to be run, the previous two screens are shown again to define a new batch, which is again followed by the performation of the simulation of that batch etc.
The final step is, depending on how the program was run, either ending and quitting the program and showing the resultant file with all the results of the simulation in case several batches have been run, or the following screen:
All information in the most recent batch is also stored in the most_recent_scenario.xls file, so that only corrections must be made next time when running the software, or when defining a new batch. To check the content of that file, open in in MS Excel.
However, some pieces of information, such as the area and biomass of forest that exists at the beginning of the scenario, can also be provided by the user in a separate file in case the afforestation program is conducted in an irregular way. This information must be provided in the user.xls file, which was created for easy, although rare access by the users. See more details here.
All program windows contain help buttons that provide information either on the model, or on the program itself.
The outputs CASMOFOR produces can be grouped into two main groups. One is found in a file named most_recent_scenario.xls that contains information about the scenarios that have been defined by the user. This is to keep log of the scenarios used for reference, and also to simplify data input the next time the program is run, because the program shows all information in this file in the input boxes so that only changes must be made by the user.
Concerning the other group, it is all those graphs and tables that CASMOFOR produces. These are accessable either from the last program window ("RESULTS"), in case the option "Running one batch" was selected, or from an MS Excel file that remains open after the running of CASMOFOR end, no matter which option of running types was selected. The name of this file is composed of "results_" plus the name of the run that you provide at the beginning of running the model (e.g., if you provide the name "oak_scenario", the name of the output file will be "results_oak_scenario".
To check the content of the above mentioned files, open them in MS Excel.
The output of the simulations include the following information for the whole period for which the simulation has been made:
For the most recent batch, accessible from the last program window ("RESULTS"), but also in the results file:
For all batches when several batches are run, accessible in the results file: totals or comparisons of
depending on whether the results of the various batches are to be compared or added up. Economic information is also displayed.
The output file contains a series of tables and graphs. These are separated on the various sheets. You can get access to the various sheets by clicking on the tabs of the sheets. The list of all sheets can be found on the very first sheet of the file, from where you can go to all other sheets directly.
Since the output is also on MS Excel sheets and graphs, they can easily be copied and pasted into other Windows-based software systems (such as MS Word). Note that the formatting of the graphs (and, to a lesser extent, of the tables) depends on the limitations of MS Excel, you may need to adjust the font type, scale, or the legend.)
For the types of results and examples of how they can be interpreted, click here.
Note that for some economic analysis, you may want to adjust the market price of a tonne of CO2 sequestered, as well as the HUF/EURO rate in the "summary" sheet of the results file.
This webpage was last modified by Zoltan Somogyi 29 June 2014.