4.1 USDA soil orders

We use the USDA soil order map of Sun et al. (in revision) with the dominant soil order per pixel. The gap filling of the missing values was done like this (see script /home/users/dgoll/ORC_data/USDA_soilorders/try_2_fill/fill_deserts.ksh):

We use the soiltext class 6 for cold desert from the input file "soils_param.nc" to fill missing values in these region with Gelisols (5). After we set cold desert, we use the level 1 of the maxvegetfrac in the input file "PFTmap.nc" to fill missing values of the remaining desert to Aridisols (2). Hereby, we use a threshold of 0.8 for identifying deserts in maxvegetfrac.

In case of the remaining missing points, mostly coast line points, ORCHIDEE will assign the value of the nearest neighbour in the routine "get_soil_orders" in src_stomate_io.f90.

Figure 1: USDA soil orders: Alfisols(1), Andisols(2), Aridisols(3), Entisols(4), Gelisols(5), Histosols(6), Inceptisols(7), Mollisols(8), Oxisols(9), Spodosols(10),Ultisols(11), Vertisols(12)

4.2 GliM lithology

We use the GliM lithology map by Hartmann & Moosdorf (2012). We account for subgrid scale heterogeneity by reading the fractional coverage of each of the 16 classes.

Figure 2: GliM lithology: the dominant type with a fractional cover of >0.51 ! alphabetical sequence of lithological classes: ! ! 0 - no dominant type ! ! 1 - evaporites ! ! 2 - ice & glaciers ! ! 3 - metamorphics ! ! 4 - no data ! ! 5 - acid plutonic rocks ! ! 6 - basic plutonic rocks ! ! 7 - intermediat plutonic rocks ! ! 8 - pyroclastics ! ! 9 - carbonate sedimentary rocks ! ! 10 - mixed sedimentary rocks ! ! 11 - siliciclastic sedimentary rocks ! ! 12 - unconsolidated sediments ! ! 13 - acid volcanic rocks ! ! 14 - basic volcanic rocks ! ! 15 - intermediate volcanic rocks ! ! 16 - water bodies

4.4 Soil shielding (Hartmann et al. (2014)

The factor corrects the weathering flux where the active zone in the soil is disconnected from the bedrock.