R scripts for statistical analysis and modelling of CO2 fluxes from portable flux chambers in the High Arctic Tundra

R scripts for statistical analysis and modelling of CO2 fluxes from portable flux chambers in the High Arctic Tundra.

The scripts are meant to be used to elaborate statistics and to visualise the results in tables and figures related to the manuscript "Identification of carbon dioxide drivers in a High Arctic tundra environment" by F. Avogadro di Valdengo et al., currently submitted for review.

In particular:

0.Descriptive Statistics & Table1_Table2_FigS2.R: Generates Table 1, Table 2 and Supplementary Figure 2 - based on the full dataset

1.Sample size & FigS1.R:  Calculates weighted mean and standard deviation with varying sample size and generates Supplementary Figure 1 - based on the full dataset

2.Model selection & TableS1_TableS2_Fig2.R:  Performs model selection and generates Tables S1 and S2 and Figure 2 (Predicted versus measured CO₂ fluxes for the best non-linear (purple) and linear (green) models for a) ER and b) GPP)

3.Test transferability & Table6S3-S6_Fig4_Fig5.R:  Tests spatial and temporal transferability and generates Figure 4 (Estimated coefficients and standard errors of the estimate for the model of a) ER and b) GPP) and Figure 5 (Estimated coefficients and standard errors of the estimate for a) ER and b) GPP models), as well as table with model coefficients and goodness-of-fit statistics

4.Random split & Fig3.R: Random partitions the dataset, fits best models on random sets, and recalculate goodness-of-fit statistics.