This repository contains all R code included in the publication "Effect of climate warming on the timing of autumn leaf senescence reverses after the summer solstice" (Zohner et al. 2023, Science)

Workflow

1. Experiment 1
   • Seasonal temperature and light manipulation experiment
   • Run Experiment1_chlorophyll.Rmd for the leaf senescence analysis
   • Run Experiment1_photosynthesis.Rmd for the photosynthesis analysis
2. Experiment 2
   • The effects of pre- and post-solstice warming on bud set timing
   • Run Experiment2_budset.Rmd for the bud set analysis
3. FluxNet analysis
   • Analyse autumn photosynthetic declines using flux tower measurements
   • Run FluxNet_analysis.Rmd for the analysis and figures
4. Harvard analysis
   • Ground-sourced phenology observations for Eastern North America (Harvard forest)
   • Run Harvard_models.Rmd to create models and figures
5. PEP725 analysis
   • Ground-sourced phenology observations for Europe (PEP725 data)
   • Step 1: 1_Data_extraction - download www.PEP725.eu data, GLDAS climate data and other info
   • Step 2: 2_Add_drivers - generate seasonal climate and photosynthesis drivers
   • Step 3: 3_Analysis - run models and generate figures
6. Remote sensing analysis
   • Analysis of satellite-derived phenology observations
   • Step 1: 1_Data_extraction - extract photoperiod info
   • Step 2: 2_Add_drivers - generate seasonal climate and photosynthesis drivers
     • EOS10 = Date when EVI last crossed 90% threshold
     • EOSstart = Date of maximum EVI curvature change rate (breakpoint)
     • EOS50 = Date when EVI last crossed 50% threshold
     • EOS85 = Date when EVI last crossed 15% threshold
   • Step 3: 3_Analysis - run models and generate figures
     • 3.1 Check sample sizes, 3.2 add preseason temperatures, 3.3 run models and 3.4 create figures
     • EOS10 folder = EOS10_SenescenceStart
     • EOSstart folder = EOSstart_VNP
     • EOS50 foler = EOS50_MidGreendown
     • EOS85 folder = EOS85_Dormancy