Supplement of Geosci
Total Page:16
File Type:pdf, Size:1020Kb
Supplement of Geosci. Model Dev., 11, 1343–1375, 2018 https://doi.org/10.5194/gmd-11-1343-2018-supplement © Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License. Supplement of LPJmL4 – a dynamic global vegetation model with managed land – Part 1: Model description Sibyll Schaphoff et al. Correspondence to: Sibyll Schaphoff ([email protected]) The copyright of individual parts of the supplement might differ from the CC BY 4.0 License. S1 Supplementary informations to the evaluation of the LPJmL4 model The here provided supplementary informations give more details to the evaluations given in Schaphoff et al.(under Revision). All sources and data used are described in detail there. Here we present ad- ditional figures for evaluating the LPJmL4 model on a plot scale for water and carbon fluxes Fig. S1 5 - S16. Here we use the standard input as described by Schaphoff et al.(under Revision, Section 2.1). Furthermore, we evaluate the model performance on eddy flux tower sites by using site specific me- teorological input data provided by http://fluxnet.fluxdata.org/data/la-thuile-dataset/(ORNL DAAC, 2011). Here the long time spin up of 5000 years was made with the input data described in Schaphoff et al.(under Revision), but an additional spin up of 30 years was conducted with the site specific 10 input data followed by the transient run given by the observation period. Comparisons are shown for some illustrative stations for net ecosystem exchange (NEE) in Fig. S17 and for evapotranspira- tion Fig. S18. Only 2 stations show a slightly better performance of LPJmL4 to NEE measurements and for evapotranspiration only 1 station, the others show a similar correlation as the simulations conducted with global climate input. 15 We use gauging station to evaluate the river discharge as an integrated measure (Fig. S19- S66). Fig. S68a compares the two evaluation data sets against each other. The comparison of the global data set from (Jung et al., 2011) to the local data (Luyssaert et al., 2007) shows that both data sets are in good agreement. The comparison of LPJmL4 against the global data set of Jung et al.(2011) on the local scale (Fig. S68b show a slightly worse match as the comparison against the local data 20 (see Fig. 4, main text). Fig. S68c compares LPJmL4 against the global data set from (Jung et al., 2011), but excluding outliers with very high GPP. That increases the match to these data to a NMSE of 0.69 and a R2 of 0.51. These comparisons show also that the comparisons on both scales are meaningful and can give a good indication how good the model can reproduce global as well as local biomass estimations by different methods. Fig. S67 and Fig. S69a give a comparison with the 25 global estimation of Carvalhais et al.(2014) for soil organic carbon resp. biomass. Additionally we have compared aboveground biomass in Fig. S69b with estimates by Liu et al.(2015). A spatial comparison of ecosystem respiration is shown in Fig. S70. Evapotranspiration comparison against MTE data (Jung et al., 2011) is shown in Fig. S71 and Fig. S72 shows a comparison of simulated fractional burnt area against remote sensing observations (GFED4: http://www.globalfiredata.org/ 30 and CCI Fire Version 4.1: http://cci.esa.int/data). Remote sensing data are also used for the evaluation of FAPAR (Fig. S73) and Albedo (Fig. S74). Sowing dates have been proved to be important to simulate crop variability (Fig. S75- S83), a comparison with MIRCA sowing dates we show in Fig. S84- S93. Table S1 gives an overview of estimates for regional field application efficiencies, showing that 35 LPJmL4 are in a similar range as other estimates. 1 Kaamanen_wetland 69.14 N / 27.3 E NEE 100 r2 = 0.31, W = 0.67, NMSE = 3.1, NME = 1.6 50 LPJmL4 fluxes 0 Euroflux and Ameriflux Data gC/m2/month −100 2000 2001 2002 2003 2004 AK−Ivotuk 68.49 N / 155.75 W Sodankyla 67.36 N / 26.64 E 2 = = = = 2 = = = = 50 r 0.17, W 0.62, NMSE 2.2, NME 1.3 r 0.68, W 0.87, NMSE 0.76, NME 0.92 100 50 0 0 gC/m2/month gC/m2/month −50 −100 2004 2005 2006 2007 2000 2001 2002 2003 2004 Flakaliden 64.11 N / 19.46 E Hyytiala 61.85 N / 24.29 E 2 = = = = 100 2 = = = = 100 r 0.53, W 0.85, NMSE 0.54, NME 0.72 r 0.78, W 0.94, NMSE 0.23, NME 0.43 50 50 0 0 −50 gC/m2/month gC/m2/month −100 −150 1997 1998 1999 2000 2001 2002 2003 1998 2000 2002 2004 Norunda 60.09 N / 17.48 E Griffin−Aberfeldy−Scotland 56.61 N / 3.8 W r2 = 0.39, W = 0.74, NMSE = 0.89, NME = 0.91 r2 = 0.6, W = 0.68, NMSE = 1.1, NME = 1 50 50 −50 −50 gC/m2/month gC/m2/month −150 −150 1996 1998 2000 2002 2004 2006 2008 −250 1998 2000 2002 2004 2006 UCI−1930_burn_site 55.91 N / 98.52 W UCI−1964_burn_site 55.91 N / 98.38 W r2 = 0.25, W = 0.65, NMSE = 1.2, NME = 0.87 r2 = 0.41, W = 0.79, NMSE = 0.67, NME = 0.73 100 50 50 0 0 gC/m2/month gC/m2/month −50 −100 −100 2002 2003 2004 2005 2002 2003 2004 2005 East_Saltoun 55.91 N / 2.86 W UCI−1981_burn_site 55.86 N / 98.48 W r2 = 0.6, W = 0.92, NMSE = 0.3, NME = 0.58 r2 = 0.38, W = 0.76, NMSE = 0.71, NME = 0.68 50 50 0 0 gC/m2/month gC/m2/month −100 −50 −100 −200 2004 2005 2006 2002 2003 2004 2005 Figure S1. Comparison of NEE fluxes with EDDY-flux measurements. 2 Lille_Valby 55.68 N / 12.08 E r2 = 0.3, W = 0.77, NMSE = 0.65, NME = 0.75 NEE 100 LPJmL4 fluxes 0 Euroflux and Ameriflux Data gC/m2/month −200 2005 2006 2007 2008 2009 Soroe−LilleBogeskov 55.49 N / 11.65 E Loobos 52.17 N / 5.74 E 2 2 r = 0.81, W = 0.95, NMSE = 0.17, NME = 0.39 100 r = 0.79, W = 0.91, NMSE = 0.46, NME = 0.67 50 50 0 0 gC/m2/month gC/m2/month −100 −100 −200 1996 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 Horstermeer 52.03 N / 5.07 E Cabauw 51.97 N / 4.93 E 100 100 r2 = 0.67, W = 0.92, NMSE = 0.37, NME = 0.58 r2 = 0.36, W = 0.72, NMSE = 1.6, NME = 1.2 50 50 0 0 −50 gC/m2/month gC/m2/month −100 −150 2004 2005 2006 2007 2004 2005 2006 2007 2008 Brasschaat 51.31 N / 4.52 E Mehrstedt 51.28 N / 10.66 E r2 = 0.75, W = 0.92, NMSE = 0.29, NME = 0.48 r2 = 0.46, W = 0.84, NMSE = 1.2, NME = 1 100 50 50 0 0 −50 gC/m2/month gC/m2/month −100 −150 2000 2002 2004 2006 2008 2004 2005 2006 2007 Gebesee 51.1 N / 10.91 E Hainich 51.08 N / 10.45 E 2 2 r = 0.15, W = 0.69, NMSE = 0.73, NME = 0.74 100 r = 0.47, W = 0.73, NMSE = 0.63, NME = 0.65 100 0 0 −100 gC/m2/month gC/m2/month −200 −300 2004 2005 2006 2007 2008 2000 2002 2004 2006 2008 Tharandt−old_spruce 50.96 N / 13.57 E Lonzee 50.55 N / 4.74 E r2 = 0.8, W = 0.83, NMSE = 0.72, NME = 0.83 r2 = 0.1, W = 0.54, NMSE = 0.79, NME = 0.8 50 0 0 −200 gC/m2/month gC/m2/month −100 −400 −200 1998 2000 2002 2004 2005 2006 2007 2008 2009 Figure S2. Comparison of NEE fluxes with EDDY-flux measurements. 3 Wetzstein 50.45 N / 11.46 E r2 = 0.66, W = 0.89, NMSE = 0.4, NME = 0.56 NEE 100 LPJmL4 fluxes 50 0 Euroflux and Ameriflux Data gC/m2/month −100 2002 2004 2006 2008 Vielsalm 50.31 N / 6 E Bily_Kriz−_Beskidy_Mountains 49.5 N / 18.54 E r2 = 0.82, W = 0.87, NMSE = 0.49, NME = 0.65 r2 = 0.59, W = 0.78, NMSE = 0.76, NME = 0.8 50 50 0 0 −50 gC/m2/month gC/m2/month −100 −150 −200 1998 2000 2002 2004 2006 2008 2000 2002 2004 2006 2008 Bily_Kriz−_grassland 49.5 N / 18.54 E Hesse_Forest−Sarrebourg 48.67 N / 7.06 E 100 r2 = 0.62, W = 0.83, NMSE = 0.74, NME = 0.77 r2 = 0.59, W = 0.78, NMSE = 0.49, NME = 0.58 50 100 0 0 gC/m2/month gC/m2/month −100 −200 2005 2006 2007 2008 2009 1998 2000 2002 2004 2006 2008 Neustift/Stubai 47.12 N / 11.32 E WI−Mature_red_pine 46.74 N / 91.17 W r2 = 0.19, W = 0.68, NMSE = 1.2, NME = 0.93 r2 = 0.34, W = 0.44, NMSE = 6.3, NME = 2.7 50 50 −50 −50 gC/m2/month gC/m2/month −150 −150 2002 2003 2004 2005 2006 2007 2008 −250 2003 2004 2005 2006 Bugacpuszta 46.69 N / 19.6 E Renon/Ritten 46.59 N / 11.43 E r2 = 0.29, W = 0.77, NMSE = 0.96, NME = 0.91 r2 = 0.51, W = 0.72, NMSE = 1.2, NME = 1 100 50 50 0 0 −50 gC/m2/month gC/m2/month −100 −150 2003 2004 2005 2006 2007 2008 2009 2000 2002 2004 2006 2008 WI−Lost_Creek 46.08 N / 89.98 W WI−Willow_Creek 45.81 N / 90.08 W 100 2 = = = = 2 = = = = r 0.14, W 0.59, NMSE 1.2, NME 0.9 100 r 0.27, W 0.51, NMSE 0.91, NME 0.77 50 0 0 −100 gC/m2/month gC/m2/month −100 −300 2001 2002 2003 2004 2005 2006 2000 2002 2004 2006 Figure S3.