In nuaquilue/QLDM: Quebec Landscape Dynamic Model

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
library(raster)
library(rasterVis)
library(tidyverse)

Introduction

The QLDM is a package to spatially explicit simulate forest landscape dynamics in Quebec province (Canada). The landscape-scale processes included are forest management (sustainable yield, clear cut, partial cut and salvage logging), wildfires and spruce budworm outbreaks. Forest stands and climatic data are distributed with the package to project Quebec's forest ecosystems dynamics under natural disturbances and climatic change scenarios.

You can learn more about the processes of the modelling framework and potential applications in

Bouchard, M., Aquilué, N., Périé, C., Lambert, M.-C. 2019. Tree species persistence under warming conditions: A key driver of forest response to climate change. Forest Ecology and Management, 442, 96-104. https://doi.org/10.1016/j.foreco.2019.03.040.

Package installation

``` {r install, eval=F} devtools::install_github("nuaquilue/QLDM")

## Model initialization

The Quebec Landscape Dynamic Model is initialized for the commercial forests of Quebec province (Canada) in 2020. Forest stands are characterized by the dominant tree species or group of species, stand age, age of maturity, soil type (`A` - Clay, `O` - Organic, `R` - Rock, `S` - Sand, `T` - Till), the status of protection, the fire regime zone, and the managment unit.

The main tree species or group of species considered are:

- `BOJ`: Yellow birch (*Betula alleghaniensis*)
- `EPN`: Black spruce (*Picea mariana*)
- `ERS`: Sugar maple (*Acer saccharum*)
- `PET`: Trembling aspen (*Populus tremuloides*)
- `SAB`: Balsam fir (*Abies balsamea*)
- `OTH.FEU.N`: Other boreal hardwoods (e.g. White birch (*Betula papyrifera*))
- `OTH.FEU.S`: Other temperate hardwoods (e.g. Basswood (*Tilia americana*), Beech (*Fagus grandifolia*), Oaks (*Quercus*), Red maple (*Acer rubrum*))
- `OTH.RES.N`: Other boreal conifers (e.g. Jack pine (*Pinus banksiana*), Tamarack (*Larix laricina*), White spruce (*Picea glauca*))
- `OTH.RES.S`: Other temperate conifers (White pine (*Pinus strobus*), Red pine (*Pinus resinosa*), Eastern hemlock (*Tsuga canadensis*), Red spruce (*Picea rubens*))

The `NonFor` category identifies the locations that are not currently forest areas.  

Included in the **QLDM** package are the data frame `landscape` that contains the forest stands characteristics and the raster `mask` that describes the spatial distribution of the forest stands.

``` {r data, eval=T, fig.width = 8, fig.height=6}
library(QLDM)
map = mask
map[!is.na(map[])] = landscape$spp
map = as.factor(map)
rat = levels(map)[[1]] # Extract the RAT
rat$spp <-  sort(unique(landscape$spp))
levels(map) <- rat
spp.color=c("darkgoldenrod1", "grey30", "darkred", "white", "darkorange4", "darkorange2", "darkolivegreen4", "darkolivegreen2", "blueviolet", "blue3")
levelplot(map, main="Forest tree species in Quebec (2020)", col.regions=spp.color, scales=list(draw=FALSE))

The historical annual mean temperature and precipitation at the stand scale are also reported in the landscape data frame. Moreover, the QLDM package includes RCP4.5 and RCP8.5 temperature and precipitation projections from the MIROC_ESM_CHEM global circulation model at 2 km of spatial resolution and 5-year of temporal resolution (temp_rcp45, prec_rcp45, temp_rcp85, and prec_rcp85, respectively).

``` {r clim, eval=T, fig.width = 8, fig.height=6} df = data.frame(cell.id=1:ncell(mask)) %>% left_join(temp_rcp85, by="cell.id") map = mask map[] = df[,2] b = stack(map) for(i in 3:ncol(temp_rcp85)){ map[] = df[,i] b = stack(b, map) } names(b) = paste("Y", seq(2020,2095,5)) levelplot(b, main="RCP8.5 Temperature projections for Quebec", scales=list(draw=FALSE), par.settings = BuRdTheme)

## Simulations of the Quebec Landscape Dynamic Model

#### Example 1

Run one single 80-year replicate of the model including forest management but not natural disturbances under default RCP 4.5 climate projections.
``` {r ex1, eval=F}
quebec.ldm(is.harvesting = T, rcp = "rcp45")

Example 2

Run one single 80-year replicate of the model including forest management but not natural disturbances under default RCP 4.5 climate projections. Change the replanif parameter to not re-calculate sustainable timber supply levels every time step.

``` {r ex2, eval=F} custom.params = default.params() custom.params$replanif = FALSE quebec.ldm(is.harvesting = T, rcp = "rcp45", custom.params = custom.params)

#### Example 3

Run one single 80-year replicate of the model including forest management and wildfires under default RCP 4.5 climate projections, and assume that forest species have no capacity to persist in non susitable environmental conditions.

``` {r ex3, eval=F}
spp.colonize.persist$persist = 0
quebec.ldm(is.wildfires = T, is.harvesting = T, rcp = "rcp45", spp.colonize.persist)

nuaquilue/QLDM documentation built on Dec. 22, 2021, 3:18 a.m.