Data platform to represent status of forests and soil (Action 2)
In the action 2 of the Climforisk project main deliverables are:
1. Biomass maps Finland
2. GPP, NPP, NEE and LAI maps for Finland
3. Thematic map for Finland indicating drought risk
4.The scientific publications used to develop methods for above mentioned products are provided in the list of References
Examples of work stages
Spatial variability of soils
Figure 1 Combination of different GIS layers: topographic maps (water bodies,
marshes and bare rocks) and soil maps (depth of mineral soil layer).
Water bodies and marshes are utilized as mask layers to block away areas
where does not excist dry conditions. Bare rock areas are used as areas
which have extremely high potential for drought due the low amount of
loose material (mineral soil or humus). Depth of mineral soil layer is
used to classify the risk of drought.
Soil depth and texture determine how much water soils can hold. Usually shallow soils with coarse texture are more drought prone than deeper or fine textured soils. Plants on the other hand need water from soils. In Finland, Scots pine tends to live on drier soils than Norway spruce, or broadleaved species.
Soil map of Finland (Lilja et al 2006) identifies soil depth and texture on coarse resolution (> 6.25 ha). However, soils are much more variable than polygon of that size. Topographical map offers much higher resolution than this and identifies open rock formations and peatlands. Using this information, we can improve soil map resolution on thin soils dominated by pines. On the other hand, we can identify areas which are unlikely suffer from drought (peatlands). See figure 1 for illustration.
Tree species distribution in Finland
|Average birch growing stock (m3/ha) in Finland.
National forest inventory data provides large scale information on the spatial distribution different tree species in Finland. Sample plot level data cannot, however, capture high resolution variability in species distribution. Instead of that, tree species volume map interpolated from plot level data provides information on the variability of species structure in the scale of ten to fifty kilometers (Fig 2). This source of information can be used as an ancillary variable in kNN-method when predicted plot-level target variables are scaled to rest of the area using remote sensing images, which themselves are limited in identifying the dominance of different conifer species. Accurate identification of tree species is important also when forest vulnerability to pest/pathogen damages is estimated, as pests/pathogens are species specific.
Deliverables and other products