rSOIL-DE

Author: Annelie Säurich
Date: 10.10.2023
Language: R Version 4.3.1

General info

The repository contains the scripts and necessary steps to implement the goals of the SOIL-DE project "Development of indicators to assess the yield capacity, utilisation intensity and vulnerability of agricultural soils in Germany". The steps are:

• Developping a general worflow to utilize the Bodenübersichtskarte 1:200.000 (BÜK200, BGR)
• Calculating soil functions and potentials to evaluate the soil
• Determining quantitative soil loss of agricultural soils
• Determining qualitative soil loss of agricultural soils

Overview over all scripts and input parameters

Developing a general worflow to utilize the BÜK200

buek200_workflow_CLC%YY.R

In the BÜK200 database, a number of similar soil profiles (BF_ID) are grouped under one universal key (GEN_ID), which is connected to the shape file. The soil profiles and associated soil properties of each soil layer can be assigned by using the land-use mapped in the CORINE Land Cover (CLC, CLMS)inventory of 2006, 2012 and 2018. Before the buek200_workflow_CLC%YY.R can be run the following data needs to be prepared:

in GIS

1. Clip the BÜK200 shape file and the ATKIS 2016 waterbodies shape file in GIS
2. Intersect BÜK200 and CORINE CLC%YYYY in GIS and filter Germany for Grassland (GL, Code 231) and Cropland (A, Code 211,221,222,242,243) and save as .shp ("bük200_clc%YY_GL" and "bük200_clc%YY_A.shp") and the columns tblZuordnu aka GEN_ID and CLC Code as .csv ("bük200_clc%YY_GL.csv" and "bük200_clc%YY_A.csv")

Now the script of the respective CLC year YY buek200_workflow_CLC%YY.R can be used. This is what happens in the script:

in R

1. Only GEN_IDs that are under Grassland or Cropland use according to CLC are considered in this workflow. All other GEN_IDs are ignored.
2. It is checked whether CLC%YY and GEN_ID land use match:
   Yes: the assignment workflow is continued.
   No: the main soil profile (BF_ID 1) is used as soil profile, never mind the land use indicated.
3. Assignment of suitable soil profile to GL or A land use:

• Take BF_ID 1 if there are matching.
• Otherwise use BF_ID with maximum area share (column FLANT_MITTELW)
• Or lastly, if the area share was the same for several BF_IDs, unsuitable soil profiles are sorted out (e.g. missing groundwater level or soil parameters) and ultimately one of these BF_IDs is randomly selected per GEN_ID. Result: "Profile_A.csv" and "Profile_GL.csv" containing the assigned BF_IDs

To relate the assigned BF_IDs to the Germany are they need to joined in GIS:

in GIS

1. Join shape "bük200_clc%YY_A.shp" and "Profile_A.csv" and add column land use 1 for A
2. Join shape "bük200_clc%YY_GL.shp" and "Profile_GL.csv" and add column land use 2 for GL
3. Merge both shape files to "A_GL_BF_ID.shp"
4. Dissolve "A_GL_BF_ID.shp" by "BF_ID" to get "clc%yy_key_BF_ID.shp"
5. Dissolve "A_GL_BF_ID.shp" by "land use" to get "Nutzung%YYYY.shp"
6. Rasterize 4. and 5. to get "clc%yy_key_BF_ID.tif" and "Nutzung%YYYY.tif"

Calculating soil functions and potentials to evaluate the soil

Input data next to BÜK200

Climate and topograhic data

• "Slope_10.tif" is the DGM10 grouped into 11 classes according to KA5
• Calculate profile curvature "curv_prof.tif" and plan curvature "curv_plan.tif" using "Slope_10.tif" in GIS
• Prepare the needed climate data using DWD.R

Water table depth
a lot of BF_IDs do not have information on water table depth
missing information are gathered using the BÜK1000 water table depth "MNGW_10.tif"
intersect of BF_ID and BÜK1000 information: "GW_Abstand_BÜK1000.csv"

General proceeding

• generation of the required grids Germany-wide Raster_prep_BÜK200.R
• create a folder for the federal state
• Clip the "clc%YY_key_BF_ID" with the federal state border in GIS
• Use BÜK200_CLC%YY.R to calculate soil functions and potential after Marks et al. (1992)
• Use SQR_CLC%YY.R to calculate the Müncheberg Soil Quality Rating (Müller et al. 2007)

Raster_prep_BÜK200.R

generating Germany-wide .tif raster per soil profile and per soil horizon depending on the soil parameters

• soil profile wise: water table depth in cm and classes and field capacity in mm and in classes, waterlogging, litter cover
• soil horizon wise: layer thickness, soil type class, bulk density, soil type, humus class, coarse soil

BÜK200_CLC%YY.R

important: replace the abbreviation of the federal state (e.g. BL <- "BB") in the whole script before starting

• before the calculation all relevant rasters are clipped to the federal state's borders
• pH value rasters are generated

Runoff-regulation function

• "ab_GesMae.tif": only upper 10 cm are used
• "ab_nr_hor.tif": number of soil horizons used for the calculation
• "ab_Hor.tif": evaluation of infiltration capacity using soil type class and coarse soil content for each soil horizon, then the weighted mean is calculated via horizon thickness
• "ab_nfk_O.tif": evaluation of field capacity and additions due to litter layer
• "ab_neig.tif": evaluation of slope gradient
• "abfl_%YYYY.tif": evaluation of soil cover/land-use and subsequent cumulative result of intermediate steps

Erosion resistance against water

• peat is not evaluated in Marks et al., but since peat has high erosion resistance against water they are evaluated as 5 (high)
• same depth evaluated as in runoff regulation
• "BEW_Hor#.tif": soil type related erosion resistance
• "korrBEW_Hor#.tif": code from humus, coarse soil and BEW used for correcting the BEW (four digits, then two digits)
• "korrBEW_Eaqua.tif": corrected soil type related erosion resistance as weighted mean, calculated via horizon thickness
• "eaqua_w.tif": influence of curvature
• "abtrag.tif": determination of mean soil removal as a function of corrected BEW, slope gradient and summer precipitation
• "erwa_%YYYY.tif": soil removal/erosion values after land-use-dependent correction

Erosion resistance against wind

• same depth evaluated as in runoff regulation and erosion resistance against water
• "ökoF/": determination of the ecological moisture level using soil type class and groundwater level
• "eros_w_oM": erosion resistance against wind for mineral soils depending on soil type, humus content and ecological moisture level
• "eros_w_hor": erosion resistance against wind including peat dependent on peat type and decomposition
• "e_wind_%YYYY": weighted mean of erosion resistance calculated via horizon thickness

Mechanical filter function

• "filter_GesMae.tif": soil thickness of all soil horizons until ground water level
• "filter_nr_hor.tif": number of soil horizons used for the calculation
• "mF_mittel.tif": evaluation using soil type class as weighted mean calculated via horizon thickness
• "mF_zus_f.tif": additions or reductions due to filtration section over groundwater level
• "mF_klimWB.tif": additions due to climatic water balance
• "MF_%YYYY.tif": final evaluation

Physical-chemical filter function

• same depth evaluated as in mechanical filter function
• "cF_mittel.tif": evaluation using soil type class as weighted mean calculated via horizon thickness
• "cF_zus_f.tif": additions or reductions due to filtration section over groundwater level
• "pcF_%YYYY.tif": final evaluation

Biotic yield potential

• "be_GesMae.tif": only upper 30 cm are used
• "be_nr_hor.tif": number of soil horizons used for the calculation
• "be_skel.tif": evaluation of coarse soil content dependent on land-use as weighted mean calculated via horizon thickness
• "be_BAOB.tif": evaluation of soil type class dependent on land-use as weighted mean calculated via horizon thickness
• "be_Naehr.tif": evaluation of nutrient status using soil type and pH-value dependent on land-use as weighted mean calculated via horizon thickness
• "be_neig.tif": evaluation of slope gradient dependent on land-use
• "be_GW.tif": evaluation of groundwater level dependent on land-use
• "be_staunass.tif": evaluation of waterlogging dependent on land-use
• "be_nFK.tif": evaluation of field capacity dependent on land-use
• "be_Temp.tif": evaluation of mean annual temperature dependent on land-use
• "be_P.tif": evaluation of annual precipitation sum dependent on land-use
• "be_erg.tif": evaluation of erosion vulnerability
• "bio_%YYYY.tif": minimal evaluation of raster cell is finale evaluation

SQR_CLC%YY.R

Basic indicators

are scored by using scoring tables
Single scores are on a quasi 5-ball scale ranking from best conditions (2) to worst (0) with possible increments of 0.5, or 0.25 in very sensitive cases
they are dependent on land-use cropland/grassland

• Slope and relief "SQR/hang.tif"
• Soil substrate
  • soil depth used for evaluation dependent on land-use
  • cropland 80 cm: "sqr_GesMae_A.tif", "sqr_nr_hor_A.tif"
  • grassland 50 cm: "sqr_GesMae_GL.tif", "sqr_nr_hor_GL.tif"
  • both together: "sqr_GesMae.tif", "sqr_nr_hor.tif"
  • "SQR/subst.tif": evaluation of substrate using soil type as weighted mean calculated via horizon
• Depth of A-horizon and depth of humic soil
  • depth of A-horizon: A-horizons and top horizons including p "sqr_depth_A.tif"
  • depth ogf humic soil: dependent on humus content "sqr_depth_GL.tif"
  • "SQR/sqr_depth.tif": comcination of cropland and grassland
• Aggregates and porosity / topsoil structure
  • evaluation after Diez&Weigelt (1997) using soil structure information
  • peat evaluated according to bulk density
  • if soil structure is NA median is used
  • "SQR/top.tif"
• Subsoil structure
  • analogous to topsoil structure
  • "SQR/sub.tif"
• Rooting depth
  • calculating effective rooting depth via soil type and bulk density (KA5) "root.tif"
  • correction according to KA5 via land-use "We.tif"
  • "SQR/r_depth.tif": final evaluation
• Profile available water
  • using field capacity [mm]
  • "SQR/PAW.tif"
• Wetness and ponding (waterlogging)
  • using groundwater level
  • "SQR/Wet.tif" "SQR/SQR_basic.tif": cumulative indicator of all basic indicators

Hazard multipliers

these properties are so critical for farming that they may limit the total soil quality

• folder "SQR/hazard" needs to be created
• Drought
  • "sqr_drought_risk.tif": scoring drought risk using climatic water balance of the vegetation period (May-August)
  • "SQR/hazard/drought": evaluation based on drought risk and land-use
• Acidification
  • "sqr_pH.tif": scoring acidification using pH-value
  • "SQR/hazard/Sauer.tif": evaluation based on acidification and land-use
• Soil depth above hard rock
  • "sqr_hardrock.tif": using previously calculated rooting depth "We.tif" as approximate measure for scoring
  • "SQR/hazard/hard_rock.tif": evaluation based on limitation free soil depth and land-use
• Coarse rocks
  • "sqr_ske.tif": scoring rock at the surface using coarse soil content
  • "SQR/hazard/coarse_rocks.tif": evaluation based on coarse soil and land-use

"SQR/SQR_hazard.tif": taking the minimal multiplier of all four hazard indicators
"SQR_%YYYY.tif": multiplying basic indicator with hazard
"SQR_5K_%YYYY.tif": grouped scores from 1 to 5

Zusam_D_Funktionen.R

combining all federal states to one Germany-wide soil function / soil potential .tif

Determining quantitative and qualitative soil loss of agricultural soils

in GIS

• "merge_cor06_%YYYY.tif": merge IMCC2006-2009 and IMCC2009-2012 with CORINE2006 into 2 bands (imcc, nutz)
• filter CORINE2006.shp by cropland (Code 211,221,222,242,243) and grassland (Code 231) and split them in a new column (crops=1, grass=2)
• "merge_lw_%YYYY": the merged rasters filtered by cropland (Code 211,221,222,242,243) and grassland (Code 231)
• "merge_cor12_%YYYY.tif": merge IMCC2012-2015 with CORINE2012 into 2 bands (imcc, nutz)
• filter CORINE2012.shp by cropland (Code 211,221,222,242,243) and grassland (Code 231) and split them in a new column (crops=1, grass=2)
• "merge_lw_%YYYY": the merged rasters filtered by cropland (Code 211,221,222,242,243) and grassland (Code 231)
• create a folder for the federal state
• clip the "merge_lw_%YYYY" with the federal state border in GIS

quantitative soil loss

in R
Loss.R

• important: replace the abbreviation of the federal state (e.g. BL <- "BB") in the whole script before starting
• "verlust_lw_%YYYY.tif": calculate new cover areas (IMCC=1) for every available time step
• save separate and cumulative (2006-2012, 2006-2015, 2006-2018) time steps
• "verlust_lw.csv": area shares of cropland and grassland

qualitative soil loss

Loss.R

• collecting of all soil functions and potentials in one folder
• "SoilFunction_%YYYY.tif_Q_%YYYY.tif": clipping soil functions and potentials with the four different "verlust_lw_%YYYY.tif"
• "Anteile_Q_%YYYY.csv": calculation of area percentages per class and function/potential
• "Anteile_Q_gesamt.csv": all four time steps together