Management scenarios
Miquel De Caceres
2024-11-08
Source:vignettes/runmodels/ManagementScenarios.Rmd
ManagementScenarios.RmdAim
The aim of this vignette is to illustrate how to use
medfateland (v. 2.5.0) to carry out simulations of
forest dynamics on a set of forest stands while evaluating a
demand-based management scenario. In particular, we will illustrate the
use of functions create_management_scenario() and
fordyn_scenario().
Preparation
Forest stands, topography and soils
All simulations in medfateland an sf object containing information of the topography, soils and vegetation of a set of target forest stands. Here we begin by loading the example data set of 100 forest stands distributed on points in the landscape:
data("example_ifn")
example_ifn## Simple feature collection with 100 features and 7 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 1.817095 ymin: 41.93301 xmax: 2.142956 ymax: 41.99881
## Geodetic CRS: WGS 84
## # A tibble: 100 × 8
## geom id elevation slope aspect land_cover_type soil
## * <POINT [°]> <chr> <dbl> <dbl> <dbl> <chr> <list>
## 1 (2.130641 41.99872) 081015_A1 680 7.73 281. wildland <df>
## 2 (2.142714 41.99881) 081016_A1 736 15.6 212. wildland <df>
## 3 (1.828998 41.98704) 081018_A1 532 17.6 291. wildland <df>
## 4 (1.841068 41.98716) 081019_A1 581 4.79 174. wildland <df>
## 5 (1.853138 41.98728) 081020_A1 613 4.76 36.9 wildland <df>
## 6 (1.901418 41.98775) 081021_A1 617 10.6 253. wildland <df>
## 7 (1.937629 41.98809) 081022_A1 622 20.6 360 wildland <df>
## 8 (1.949699 41.9882) 081023_A1 687 14.4 324. wildland <df>
## 9 (1.96177 41.98831) 081024_A1 597 11.8 16.3 wildland <df>
## 10 (1.97384 41.98842) 081025_A1 577 14.6 348. wildland <df>
## # ℹ 90 more rows
## # ℹ 1 more variable: forest <list>To speed-up simulations in this vignette we select only stands 31 to 40:
example_subset <- example_ifn[31:40, ]
example_subset## Simple feature collection with 10 features and 7 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 1.901727 ymin: 41.96974 xmax: 2.022399 ymax: 41.97083
## Geodetic CRS: WGS 84
## # A tibble: 10 × 8
## geom id elevation slope aspect land_cover_type soil
## <POINT [°]> <chr> <dbl> <dbl> <dbl> <chr> <list>
## 1 (1.901727 41.96974) 081047_A1 478 12.0 259. wildland <df>
## 2 (1.925861 41.96997) 081048_A1 540 16.4 109. wildland <df>
## 3 (1.937928 41.97008) 081049_A1 636 7.65 263. wildland <df>
## 4 (1.949995 41.97019) 081050_A1 722 18.9 204. wildland <df>
## 5 (1.962062 41.9703) 081051_A1 763 9.63 115. wildland <df>
## 6 (1.974129 41.97041) 081052_A1 642 9.32 156. wildland <df>
## 7 (1.986197 41.97052) 081053_A1 640 17.0 326. wildland <df>
## 8 (1.998264 41.97062) 081054_A1 552 10.6 88.7 wildland <df>
## 9 (2.010331 41.97073) 081055_A1 593 5.73 265. wildland <df>
## 10 (2.022399 41.97083) 081056_A1 601 5.26 84.8 wildland <df>
## # ℹ 1 more variable: forest <list>Climate forcing
In this example we will use, for simplicity, the same weather data for all stands, but normal applications should address spatial variation of weather. In the following code we prepare a three-year meteorological data in two blocks (data frames), using the example weather data provided in medfate package:
Management prescriptions
Management scenarios require classifying forest stands into management units. Each management unit can be interpreted as a set of stands that are managed following the same prescriptions. Management units can be arbitrarily defined, but here we will define them on the basis of dominant tree species. The following code allows determining the dominant tree species in each of the 10 forest stands:
example_subset$dominant_tree_species <- sapply(example_subset$forest,
stand_dominantTreeSpecies, SpParamsMED)And the result is:
example_subset$dominant_tree_species## [1] "Pinus halepensis" "Pinus nigra" "Quercus faginea" "Pinus nigra"
## [5] "Pinus nigra" "Pinus halepensis" "Pinus nigra" "Pinus nigra"
## [9] "Pinus nigra" "Pinus nigra"Package medfateland includes a table with default prescription parameters for a set of species. This is loaded using:
data("defaultPrescriptionsBySpecies")The columns of this data frame are the same as the parameter names
required by function defaultManagementFunction() of package
medfate:
names(defaultPrescriptionsBySpecies)## [1] "Name" "SpIndex" "type"
## [4] "targetTreeSpecies" "thinning" "thinningMetric"
## [7] "thinningThreshold" "thinningPerc" "minThinningInterval"
## [10] "yearsSinceThinning" "finalMeanDBH" "finalPerc"
## [13] "finalPreviousStage" "finalYearsBetweenCuts" "finalYearsToCut"
## [16] "plantingSpecies" "plantingDBH" "plantingHeight"
## [19] "plantingDensity" "understoryMaximumCover"whereas the rows correspond to species or species groups, whose names are:
defaultPrescriptionsBySpecies$Name## [1] "Abies/Picea/Pseudotsuga spp."
## [2] "Betula/Acer spp."
## [3] "Castanea sativa"
## [4] "Eucalyptus spp."
## [5] "Fagus sylvatica"
## [6] "Fraxinus spp."
## [7] "Juniperus thurifera"
## [8] "Cupressus spp."
## [9] "Pinus halepensis"
## [10] "Pinus nigra"
## [11] "Pinus pinaster"
## [12] "Pinus pinea"
## [13] "Pinus radiata"
## [14] "Pinus sylvestris"
## [15] "Pinus uncinata"
## [16] "Chamaecyparis lawsoniana"
## [17] "Thuja spp."
## [18] "Larix spp."
## [19] "Quercus ilex"
## [20] "Quercus faginea"
## [21] "Quercus suber"
## [22] "Quercus robur/petraea/rubra/canariensis"
## [23] "Quercus pyrenaica/pubescens"
## [24] "Platanus spp."
## [25] "Populus spp."
## [26] "Salix spp."
## [27] "Otras frondosas"To specify the management unit for stands, we first define a column
management_unit with missing values and then assign the
corresponding row number of defaultPrescriptionsBySpecies
for stands dominated by each of the three species.
example_subset$management_unit <- NA
example_subset$management_unit[example_subset$dominant_tree_species=="Pinus halepensis"] <- 9
example_subset$management_unit[example_subset$dominant_tree_species=="Pinus nigra"] <- 10
example_subset$management_unit[example_subset$dominant_tree_species=="Pinus sylvestris"] <- 14
example_subset[,c("id", "dominant_tree_species", "management_unit")]## Simple feature collection with 10 features and 3 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 1.901727 ymin: 41.96974 xmax: 2.022399 ymax: 41.97083
## Geodetic CRS: WGS 84
## # A tibble: 10 × 4
## id dominant_tree_species management_unit geom
## <chr> <chr> <dbl> <POINT [°]>
## 1 081047_A1 Pinus halepensis 9 (1.901727 41.96974)
## 2 081048_A1 Pinus nigra 10 (1.925861 41.96997)
## 3 081049_A1 Quercus faginea NA (1.937928 41.97008)
## 4 081050_A1 Pinus nigra 10 (1.949995 41.97019)
## 5 081051_A1 Pinus nigra 10 (1.962062 41.9703)
## 6 081052_A1 Pinus halepensis 9 (1.974129 41.97041)
## 7 081053_A1 Pinus nigra 10 (1.986197 41.97052)
## 8 081054_A1 Pinus nigra 10 (1.998264 41.97062)
## 9 081055_A1 Pinus nigra 10 (2.010331 41.97073)
## 10 081056_A1 Pinus nigra 10 (2.022399 41.97083)The stands with missing values of management_unit (here
only one) will not be managed during simulations.
Demand-based management scenario
Management scenarios are defined using function
create_management_scenario(). Three different kinds of
scenarios are allowed. Two of them being demand-based, meaning that
management is constrained by a prescribed wood demand. The remaining one
allows actual management to freely the interplay of forest dynamics and
management prescriptions. Here we will define a demand-based scenario
with fixed demand values for all years of the simulation. The demand
will normally depend on the species available in the target stands and
the area they represent. In this example, we will require an annual
extraction of 2300 m3 of Pinus nigra or P. sylvestris and 1000 m3 of P.
halepensis:
scen <- create_management_scenario(defaultPrescriptionsBySpecies,
c("Pinus nigra/Pinus sylvestris" = 1300,
"Pinus halepensis" = 500))Note that we included the data frame
defaultPrescriptionBySpecies in our call to
create_management_scenario(). This allows having the
sylvicultural prescriptions and other management parameters in a single
object. Simulations will assign management parameters using
management_unit column and the information in the
prescription data frame.
The scenario object is a list with the following elements.
names(scen)## [1] "scenario_type" "annual_demand_by_species"
## [3] "extraction_rate_by_year" "units"The first one specifies the type of scenario, which in this case is based on a fixed input demand:
scen$scenario_type## [1] "input_demand"The next element contains the demand values we entered:
scen$annual_demand_by_species## Pinus nigra/Pinus sylvestris Pinus halepensis
## 1300 500The following element is NULL in this case, since it is used to specify demand-based scenarios where actual demand depends on observed growth and a desired rate of extraction:
scen$extraction_rate_by_year## NULLFinally, element units contains the data frame of
management units, which in our case is
defaultPrescriptionBySpecies.
Normally, we will not need to modify management parameters, but here we will make thinning operations more likely for all three species by lowering the basal area threshold that triggers them:
scen$units[c(9,10,14),"thinningThreshold"] <-20Area represented by forest stands
Before running simulations, it is necessary to specify the area (in ha) that each forest stand represents. This is important, because all wood volumes are defined at the stand level in units of m3/ha. Hence, we need to multiply these values for the actual area that the stand represents, in order to know how much of the demand is fulfilled
In this example, we will assume a constant area of 100 ha for all stands:
example_subset$represented_area_ha <- 100Carrying out simulations
We are now ready to launch the simulation of the management scenario.
This is done using a call to function fordyn_scenario(). As
other simulation functions of medfateland, we need to
supply the sf object, a table of species parameter values
and the source of weather information. We also specify the management
scenario and set parallelize = TRUE to speed-up
calculations. This last parameter will also be important in real-case
simulations.
fs_12 <- fordyn_scenario(example_subset, SpParamsMED, meteo = meteo_01_02,
management_scenario = scen,
parallelize = TRUE)## ## ── Simulation of a management/fire scenario with fordyn ────────────────────────## ℹ Checking sf input## ✔ Checking sf input [8ms]## ## ℹ Checking meteo object input## ✔ Checking meteo object input [13ms]## ## ── Scenario parameters ──## ## • Number of stands: 10## • Represented area: 1000 ha## • Number of years: 2## • Management scenario type: input_demand## • Adding column 'management_arguments'## • Default volume function## • Initial volume: 66449 m3## • Seed dispersal process included.## ## ── Simulation ──## ## ── [ Year 2001 (1/2) ]## • Demand (incl. offset): 1800 m3## • Determining available volumes and final cuts## • Demand (after final cuts): 1800 m3## • Determining thinning operations## • Seed bank dynamics and seed dispersal...## • Calling fordyn_spatial...## ℹ Checking sf input## ✔ Checking sf input [9ms]## ## ℹ Checking meteo object input## ✔ Checking meteo object input [12ms]## ## ℹ Preparing data for parallelization## ✔ Preparing data for parallelization [14ms]## ## ℹ Launching parallel computation (cores = 7; chunk size = 2)## ✔ Launching parallel computation (cores = 7; chunk size = 2) [15.1s]## ## ℹ Retrieval of results## ✔ Retrieval of results [30ms]## ## ✔ No simulation errors detected## • Final volume: 64108 m3## ## ── [ Year 2002 (2/2) ]## • Demand (incl. offset): 2762 m3## • Determining available volumes and final cuts## • Demand (after final cuts): 2762 m3## • Determining thinning operations## • Seed bank dynamics and seed dispersal...## • Calling fordyn_spatial...## ℹ Checking sf input## ✔ Checking sf input [7ms]## ## ℹ Checking meteo object input## ✔ Checking meteo object input [14ms]## ## ℹ Preparing data for parallelization## ✔ Preparing data for parallelization [13ms]## ## ℹ Launching parallel computation (cores = 7; chunk size = 2)## ✔ Launching parallel computation (cores = 7; chunk size = 2) [11s]## ## ℹ Retrieval of results## ✔ Retrieval of results [26ms]## ## ✔ No simulation errors detected## • Final volume: 62248 m3## ## ── Arranging output ──## ## • Tree/shrub tables## • Wood volume tableGiven that it is a long process, the function produces a lot of
output (this can be turned of if progress = FALSE). First,
the scenario parameters are presented, including the scenario type, the
specified demand and the number of forest stands in each management
unit. Then, the output is produced for each year, where the function
first decides and reports how many forest stands will be simulated and,
after these are completed, it summarizes the results.
Function fordyn_scenario() returns a list whose elements
are:
names(fs_12)## [1] "result_sf" "result_volumes" "result_volumes_spp"
## [4] "result_volumes_demand" "next_demand" "next_sf"The first four elements, those named result_* contain
the actual simulation results, whereas the last two elements are used
for subsequent simulations (see next section).
Stand-level results are available in element result_sf.
The column names of this sf object should be easy to
interpret if you have experience with functions fordyn() or
fordyn_spatial():
fs_12$result_sf## Simple feature collection with 10 features and 8 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 1.901727 ymin: 41.96974 xmax: 2.022399 ymax: 41.97083
## Geodetic CRS: WGS 84
## # A tibble: 10 × 9
## geometry id tree_table shrub_table dead_tree_table
## <POINT [°]> <chr> <list> <list> <list>
## 1 (1.901727 41.96974) 081047_A1 <tibble [51 × 11]> <tibble> <tibble>
## 2 (1.925861 41.96997) 081048_A1 <tibble [45 × 11]> <tibble> <tibble>
## 3 (1.937928 41.97008) 081049_A1 <tibble [16 × 11]> <tibble> <tibble>
## 4 (1.949995 41.97019) 081050_A1 <tibble [47 × 11]> <tibble> <tibble>
## 5 (1.962062 41.9703) 081051_A1 <tibble [36 × 11]> <tibble> <tibble>
## 6 (1.974129 41.97041) 081052_A1 <tibble [56 × 11]> <tibble> <tibble>
## 7 (1.986197 41.97052) 081053_A1 <tibble [62 × 11]> <tibble> <tibble>
## 8 (1.998264 41.97062) 081054_A1 <tibble [71 × 11]> <tibble> <tibble>
## 9 (2.010331 41.97073) 081055_A1 <tibble [39 × 11]> <tibble> <tibble>
## 10 (2.022399 41.97083) 081056_A1 <tibble [64 × 11]> <tibble> <tibble>
## # ℹ 4 more variables: dead_shrub_table <list>, cut_tree_table <list>,
## # cut_shrub_table <list>, summary <list>Another important element of the results is
result_volumes, which contains several volumes statistics
(in m3) summarizing what happened each year of the simulation.
First, we can inspect for each year the volume corresponding to the initial and final standing stock, the forest growth and the extracted wood:
fs_12$result_volumes[,1:7]## # A tibble: 2 × 7
## Year initial growth mortality extracted final cumulative_growth
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2001 66449. 902. 51.9 3190. 64108. 902.
## 2 2002 64108. 789. 48.0 2602. 62248. 1691.The same figures can be inspected, but corresponding to those species for which demand has been defined:
fs_12$result_volumes[,c(1,8:11)]## # A tibble: 2 × 5
## Year cumulative_extraction initial_target growth_target mortality_target
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2001 3190. 62682. 878. 47.6
## 2 2002 5792. 62675. 773. 46.9Finally, we can display for each step what was the nominal demand (according to the input), the actual demand (once the offset of previous years was included), as well as the cumulative nominal demand and cumulative extracted volumes. While strong disagreements can exist between demand and extracted wood at the annual level, the cumulative columns mentioned are important to check whether simulations fulfilled the required demand in the long term.
fs_12$result_volumes[,c(1, 12:16)]## # A tibble: 2 × 6
## Year extracted_target final_target nominal_demand demand_offset actual_demand
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2001 838. 64108. 1800 0 1800
## 2 2002 2602. 62248. 1800 962. 2762.Continuing a previous simulation
As mentioned above, other elements of the output of
fordyn_scenario() allow conducting a new scenario
simulation starting at the point where the previous one finished. The
element next_sf contains the sf object
corresponding to the final state of the simulation:
fs_12$next_sf## Simple feature collection with 10 features and 12 fields
## Geometry type: POINT
## Dimension: XY
## Bounding box: xmin: 1.901727 ymin: 41.96974 xmax: 2.022399 ymax: 41.97083
## Geodetic CRS: WGS 84
## # A tibble: 10 × 13
## geom id elevation slope aspect land_cover_type soil
## * <POINT [°]> <chr> <dbl> <dbl> <dbl> <chr> <list>
## 1 (1.901727 41.96974) 081047_A1 478 12.0 259. wildland <soil>
## 2 (1.925861 41.96997) 081048_A1 540 16.4 109. wildland <soil>
## 3 (1.937928 41.97008) 081049_A1 636 7.65 263. wildland <soil>
## 4 (1.949995 41.97019) 081050_A1 722 18.9 204. wildland <soil>
## 5 (1.962062 41.9703) 081051_A1 763 9.63 115. wildland <soil>
## 6 (1.974129 41.97041) 081052_A1 642 9.32 156. wildland <soil>
## 7 (1.986197 41.97052) 081053_A1 640 17.0 326. wildland <soil>
## 8 (1.998264 41.97062) 081054_A1 552 10.6 88.7 wildland <soil>
## 9 (2.010331 41.97073) 081055_A1 593 5.73 265. wildland <soil>
## 10 (2.022399 41.97083) 081056_A1 601 5.26 84.8 wildland <soil>
## # ℹ 6 more variables: forest <list>, dominant_tree_species <chr>,
## # management_unit <dbl>, represented_area_ha <dbl>,
## # management_arguments <list>, state <list>On the other hand, in demand-based scenarios there may be demand offsets that need to be carried on to the next simulations:
fs_12$next_demand## $offset
## Pinus nigra/Pinus sylvestris Pinus halepensis
## -37.01056 197.63592
##
## $last_growth
## [1] 773.3385In addition to the demand offset for each species or species group,
note that next_demand also contains information about the
last growth. This is necessary for scenarios where demand is modulated
depending on intended extraction rates and past growth.
If we want to carry on simulations for an extra time period (one year
in our example), we can simply call fordyn_scenario() along
with the result of the previous simulation instead of the original
sf object:
fs_3 <- fordyn_scenario(fs_12, SpParamsMED, meteo = meteo_03,
management_scenario = scen,
parallelize = TRUE)## ## ── Simulation of a management/fire scenario with fordyn ────────────────────────## ℹ Recovering previous run## ✔ Recovering previous run [6ms]## ## ℹ Checking sf input## ✔ Checking sf input [12ms]## ## ℹ Checking meteo object input## ✔ Checking meteo object input [13ms]## ## ── Scenario parameters ──## ## • Number of stands: 10## • Represented area: 1000 ha## • Number of years: 1## • Management scenario type: input_demand## • Default volume function## • Initial volume: 62248 m3## • Seed dispersal process included.## ## ── Simulation ──## ## ── [ Year 2003 (1/1) ]## • Demand (incl. offset): 1961 m3## • Determining available volumes and final cuts## • Demand (after final cuts): 1961 m3## • Determining thinning operations## • Seed bank dynamics and seed dispersal...## • Calling fordyn_spatial...## ℹ Checking sf input## ✔ Checking sf input [7ms]## ## ℹ Checking meteo object input## ✔ Checking meteo object input [13ms]## ## ℹ Preparing data for parallelization## ✔ Preparing data for parallelization [14ms]## ## ℹ Launching parallel computation (cores = 7; chunk size = 2)## ✔ Launching parallel computation (cores = 7; chunk size = 2) [10.2s]## ## ℹ Retrieval of results## ✔ Retrieval of results [19ms]## ## ✔ No simulation errors detected## • Final volume: 63009 m3## ## ── Arranging output ──## ## • Tree/shrub tables## • Wood volume tableNote that in this case, the initial output makes explicit that a previous simulation is continued.