###########################################
Description
###########################################
We performed simulations over sites with the global terrestrial model state-of-the-art
QUINCY for sites covering the high-Arctic. These simulations server to improve 
quantifications of impacts caused by nitrogen mobilized following the thaw of 
permafrost, both for vegetation growth and soil processes. In a model version that 
was extended for a better representation of high latitudes, we performed sets of 
simulations to seperate effects that arise from increased nutrient release from 
thawing of permafrost soil, changes in the physical climate, as well as atmospheric 
CO2 fertilization. In addition to this, we constructed GPP estimates from 
eddy-covariance towers, as well as seasonalities of thaw depths using soil 
temperature data.

###########################################
Methods
###########################################
Modelled data was simulated with the fully coupled QUINCY (QUantifying Interactions 
between terrestrial Nutrient CYcles and the climate system) model (see model code). 
We extended the model with important high-latitude processes (soil freezing, snow, 
dynamic rooting depths). We used University of East Anglia Climatic Research Unit 
Japanese Reanalysis (CRU-JRA; Harris, 2019) atmospheric forcing to drive the model 
for 1901 to 2018, using only the 1960-2018 time frame for our analysis. We conducted 
three sets of simulations. climate+withoutpermafrostCNP considers changes in climate, 
but initialising carbon and nutrients contents to exponentially decrease with depth, 
as in the standard model. By doing this, C, N and P contents at depth are close to 
zero, thus excluding any potential fertilization effect linked to a deepening active 
layer. The second set of simulations was again driven by changing climate (climate), 
but this time also considering the release of carbon and nutrient pools from 
previously permanently-frozen layers, i.e. release from the permafrost. The third 
set of simulations additionally considered the impact of increasing atmospheric CO2 
levels on vegetation dynamics and carbon cycle processes (climate+CO2).

Model Code is available under the DOI: 10.17871/quincy-model-2019

###########################################
Data Folders 
###########################################
- Annual_data_1960_2018.zip

Annual_data_1960_2018.zip contains annual mean timeseries of important ecosystem 
variables for the time period 1960-2018. The filename format is the following:

{model variable}_yearly_{experiment name}

The experiment names mean the following simulation series:
-PF: climate+withoutpermafrostCNP
-climate: climate
-co2: climate+co2

Columns are the fluxnet site abbreviations
Rows are the years from 1960-2018

File units:
Tsoil_30cm_yearly_{experiment name} :                             [°C]
Tair_30cm_yearly_{experiment name} :                              [°C]
resp_yearly_{experiment name} :                                   [g C m-2]
plantuptake_n_yearly_{experiment name} :                          [g N m-2 yr-1]
PF_respirationyearly_{experiment name} (deep-soil respiration):   [g C m-2 yr-1]
NEE_yearly_{experiment name}: 						[g C m-2 yr-1]
n2o_yearly_{experiment name}:                                     [g N m-2 yr-1]
N_min_80cm_yearly_{experiment name} (deep-soil N mineralization): [g N m-2 yr-1]
GPP_yearly_{experiment name}:                                     [g C m-2 yr-1]


- Seasonal_data_1960_2018.zip

Annual_data_1960_2018.zip contains weekly mean data for 1950-1970 or 1998-2018.
The filename format is the following:

model_PF_seasonality_1998_2018                                     [m]
model_GPP_seasonality_{}                                           [micromol C m-2 s-1]
model_nitrification_seasonality_{}                                 [g N week-1]
model_denitrification_seasonality_{}                               [g N week-1]
model_n2o_seasonality_{}                                           [g N week-1]
model_DeepSoilN_seasonality_1998_2018                              [g N week-1]

- eddyfluxcovariance_data.zip

eddyfluxcovariance_data.zip contains eddy covariance-based GPP seasonality for given
time periods collected at Samoylov and Cherskiy. 

observation_GPP_seasonality_Samoylov_2012_2017                      [micromol C m-2 s-1]
observation_GPP_seasonality_Cherskiy_2015_2018                      [micromol C m-2 s-1]