To simulate the Konza irrigation experiment, we used meteorological forcing data, environmental information (e.g., soil texture), and functional traits for Andropogon gerardii and Panicum virgatum (collated from author measurements and TRY) to calibrate and operate four models from 1991-2012. The goal was to evaluate the models’ ability to simulate the community shift and accompanying changes in ANPP. For these simulations, two PFTs were parameterized based on empirical information about functional traits for A. gerardii and P. virgatum and models were tuned to match observations from non-irrigated plots (i.e., controls). This led to slightly different starting PFT abundances among models when irrigated simulations began. The following protocol was used to run all models for model simulation exercise. Spin up: Spin-up to equilibrium (1000-2000 years) using standard spin-up from model. Plant functional type (PFT) is C4 grassland or similar cover type. Starting C and biomass pools should stabilize near values in Table A1. Industrial period: Run models from 1860 to 1990 with random annual draws from met data. Site is burned every 3 years at the beginning of the growing season through 1990 (very low mortality of grass individuals). PFT throughout industrial period is C4 grassland, dominant species is Andropogon gerardii (Table A1) Model calibration: Calibrate model using ambient ANPP, forcing, LAI, species relative cover (see Collins et al. 2012 for control data), soil moisture data, and additional site-level biotic and abiotic information (Table A1). Simulation of irrigation experiment: Run models in “site-mode.” Perform two simulations from 1991-2012: one with ambient precipitation forcing data, and one with irrigation forcing data. Use historical CO2 and N deposition data within model. Site is burned every year in March starting in 1991. Soil moisture observations - Volumetric soil moisture was measured hourly from 2007-2012 from 0-15 cm in soil (integrated across this range). Data to be used during model calibration Interspecific competition – If the model allows for competition, parameterize two species: A. gerardii and Panicum virgatum. A. gerardii is a [relatively] drought-tolerant perennial C4 grass. Under low water and nutrient conditions, A. gerardii is an excellent competitor for light, but competitive ability becomes weaker under higher levels of water and/or nitrogen. P. virgatum undergoes water stress under lower soil moisture conditions but is a strong competitor for light when water and nutrients are abundant. See Table A1 for species characteristics (below). Outputs: Please output daily values of the following variables (finer temporal resolution is better): ● NPP (g C m-2 d-1) - site level observations include reproductive biomass so please report NPP or biomass increment+reproductive biomass ● Stem growth (g C m-2 d-1) ● Leaf growth (g C m-2 d-1) ● Root growth (g C m-2 d-1) ● PhenoGrass: LAI ● fAPAR ● If model includes sources of biomass not captured in NPP observations, output these components (g C m-2 d-1) ○ e.g., root exudates, root turnover, reproductive biomass ● Emissions from fire (g C m-2 d-1) ● GPP (g C m-2 d-1) ● NEE (g C m-2 d-1) ● Autotrophic respiration (g C m-2 d-1) ● Transpiration (mm H20 d-1) ● Runoff (mm H20 d-1) ● Deep and shallow soil moisture (%; 0-15 cm and >15 cm). Soil moisture observations integrated from 0-15 cm. Table A1. Values used for parameterizing models. Variable units value source comments latitude degrees N 39.11 longitude degrees E -96.61 elevation m 311 met var measurement height m 3 experiment start year 1991 experiment end year 2012 Site history landuse history – plant cover na tallgrass prairie, burned annually starting in 1991 (additional rows for land use history – add as necessary) na na na na landuse history – fertiliser addition na none disturbance history – fire na Annual spring burns 1991-present na pre 1991, site burned every three years na disturbance history – hurricane/severe wind event na planting date na na grazing none Site initialisation PFT na C4 grassland PFT cover % 100 tree age at beginning of experiment years 0 vertical structure na Both Andropogon gerardii and Panicum virgatum are tall "canopy" species planting density Individuals m-2 na stand density at start of experiment Individuals m-2 51.5 # of dominant grass individuals per m2 in control plots in 1992 Mean (sd) dbh of each stem at start of experiment (m2 ha-1) cm no data Dates of harvesting during experiment na September Leaf (& non-woody shoot) biomass C g m-2 148 Wood biomass C (inc. coarse roots) g m-2 0 Fine root biomass C g m-2 419.3 from soil cores 0-30 cm, Wilcox et al. 2016 JGR Leaf (& non-woody shoot) biomass N g m-2 3.18936 0.97% N leaf biomass times aboveground biomass from Wilcox et al. 2016 JGR; 328.8 ANPP * 0.0097 Wood biomass N (inc. coarse roots) g m-2 0 Fine root biomass N g m-2 7.26882 0.78% N leaf biomass times fine root biomass from Wilcox et al. 2016 JGR; 931.9 ANPP * 0.0078 coarse woody debris C g m-2 0 soil C g m-2 9000 total C from 0-25 cm soil C g m-2 450 Active C - 5% of total soil C g m-2 5344 Slow C - 60% of total soil C g m-2 3206 Passive C - 35% of total soil N g m-2 625 Total N 0-25 cm soil N g m-2 155 Total N 0-5 cm g m-2 g m-2 Root mass in top 10 cm % 52 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass in 10-20 cm % 24 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass in 20-35 cm % 11 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass in 35-50 cm % 6 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass in 50-75 cm % 4 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass in 75-100 cm % 2 Used exponential decay function for grass PFT. Jackson et al. 1996. Root mass below 100 cm % 1 Used exponential decay function for grass PFT. Jackson et al. 1996. root:shoot in control plots na 1.1 From BNPP:ANPP measurements in Wilcox et al. 2016 JGR root:shoot in irrigated plots na 0.91 From BNPP:ANPP measurements in Wilcox et al. 2016 JGR Soil properties Field capacity % volumetric 35 Derived from soil moisture data within the experiment. Data can be found here: 10.6073/pasta/3037d44d496df3d12fbe9668c7d618a6 Wilting point % volumetric 12 Derived from soil moisture data within the experiment. Data can be found here: 10.6073/pasta/3037d44d496df3d12fbe9668c7d618a6 Bulk Density g cm-1 1.01 Sand content % 7.1 average of A1,2,3 horizons, 0-32 cm depth, Tuttle soils from Knapp et al. 1998 (grassland dynamics book) - Silt content % 50.1 average of A1,2,3 horizons, 0-32 cm depth, Tuttle soils from Knapp et al. 1998 (grassland dynamics book) - Clay content % 42.8 average of A1,2,3 horizons, 0-32 cm depth, Tuttle soils from Knapp et al. 1998 (grassland dynamics book) - sd – standard deviation Temp BlDc – temperate broadleaf deciduous Species level information Andropogon gerardii Vcmax,25 micromol CO2 m-2 s-1 34 Amax micromol CO2 m-2 s-1 21.6 Proportion roots in top 50 cm % 93 based on exponential decay function Proportion individuals surviving fire % 99 Water stress threshold for leaf abscission %AWC Operates better under low soil moisture than P. virgatum Stomatal conductance milli mol m-2 s-1 136 Specific leaf area mm2 mg-1 15.7 KW replaced the TRY number (19.6) with one calculated from data from Konza prairie (wilcox, unpublished) Rooting depth m 0.508 Leaf nitrogen (N) content per leaf area g m-2 0.625 Leaf nitrogen (N) content per leaf dry mass mg/g 10.86633 root:shoot (BNPP:ANPP) - 1.1 BNPP:ANPP in A. gerardii dominated control plots. Data from Wilcox et al. 2016 JGR Biogeosciences Reproductive effort % of total plant biomass that is flowering stock 10.7 From Knapp 1984; Also, A. gerardii typically reproduces clonally Root turnover % year-1 50 based on relationships between A. gerardii cover and root turnover - estimated root turnover at 100% A.gerardii cover from trendline. Data from Wilcox et al. 2016 JGR Biogeosciences Max leaf height cm 74 Does not include flowering stalks; Wilcox unpublished data Panicum virgatum Vcmax,25 micromol CO2 m-2 s-1 34 Amax micromol CO2 m-2 s-1 31.7 Proportion roots in top 50 cm % 93 Proportion individuals surviving fire % 99 Water stress threshold for leaf abscission %AWC Operates worse under low soil moisture than A. gerardii Stomatal conductance milli mol m-2 s-1 304 Specific leaf area mm2 mg-1 22.90292 Rooting depth m 0.3048 Leaf nitrogen (N) content per leaf area g m-2 0.88284 Leaf nitrogen (N) content per leaf dry mass mg/g 17.22296 root:shoot (BNPP:ANPP) - 0.72 based on relationships between P. virgatum cover and root:shoot - estimated root:shoot at 100% P.virgatum cover from trendline. Data from Wilcox et al. 2016 JGR Biogeosciences Reproductive effort % of total plant biomass that is flowering stock 28.6 From Knapp 1984; Also, P. virgatum typically reproduces clonally Root turnover % year-1 30.9 based on relationships between P. virgatum cover and root turnover - estimated root turnover at 100% P.virgatum cover from trendline. Data from Wilcox et al. 2016 JGR Biogeosciences Max leaf height cm 88 Does not include flowering stalks; Wilcox unpublished data  Table A2. Forcing units used to drive models Variable Unit and derivation Air temperature degC Soil temperature degC Relative humidity Percentage of the saturated vapor partial pressure VPD: VDEF, Pa (calculated from saturation vapor pressure (es): 0.6108* exp(17.27*T/(T+237.3)) and actual vapor pressure: RH/100*es. VPD=es-ea) Precipitation mm/hour Solar radiation W/m2 Wind speed m/s References cited: Knapp, A.K., Briggs, J.M., Hartnett, D.C. and Collins, S.L. eds., 1998. Grassland dynamics: long-term ecological research in tallgrass prairie. New York: Oxford University Press. Jackson, R.B., Canadell, J., Ehleringer, J.R., Mooney, H.A., Sala, O.E. and Schulze, E.D., 1996. A global analysis of root distributions for terrestrial biomes. Oecologia, 108, pp.389-411. Wilcox, K.R., Blair, J.M. and Knapp, A.K., 2016. Stability of grassland soil C and N pools despite 25 years of an extreme climatic and disturbance regime. Journal of Geophysical Research: Biogeosciences, 121(7), pp.1934-1945.

# Irrigation simulation output [https://doi.org/10.5061/dryad.m37pvmd80](https://doi.org/10.5061/dryad.m37pvmd80) The data here describe empirical observations and output from four model simulations of a long-term irrigation experiment at Konza Prairie Biological Station near Manhattan, KS, USA. ## Description of the data and file structure ANPP data: There are three columns contained in the file: year=calendar year; anpp\_gm2=aboveground net primary productivity including leaf, stem, and seed productivity in grams per meter squared per year; source=model name or empirical observation indicator. ## Sharing/Access information Links to other publicly accessible locations of the data: * Raw empirical data can be obtained from the doi: 10.6073/pasta/3037d44d496df3d12fbe9668c7d618a6 ## Code/Software na

To demonstrate current capabilities in modeling herbaceous ecosystems, we selected four different process-based models that vary in their representation of community change from no community representation to vegetation demographic models. These models were used to simulate a long-term irrigation experiment at a US tallgrass prairie (Konza Prairie Biological Station) following a standardized simulation protocol. Specifically, we were interested in how model output under a monotonic increase in water availability matched up to experimental findings of (1) herbaceous plant community change and (2) aboveground net primary productivity before and after the plant community change. The results of this simulation are included here.