Data from: The opposing effects of shortened days and cooling temperatures on tundra plant senescence
Data files
Jul 31, 2026 version files 169.60 KB
-
allmeas_flo.csv
3.36 KB
-
allmeas.csv
48.92 KB
-
Chamber_conditions.xlsx
12.33 KB
-
ChamberAnalysisCleaned.R
31.76 KB
-
ChamberModelComparisonTable.xlsx
11.27 KB
-
chambers_assignments_all_readable.csv
28.37 KB
-
README.md
4.33 KB
-
Seed_Registry_Readable.csv
29.26 KB
Abstract
Plant responses to global change are well-studied in spring and early summer but less so in late summer, particularly in high-latitude systems. It remains unclear whether senescence is driven more by temperature or photoperiod, and whether species respond similarly under late-season warming. We used growth chambers to test how photoperiod and late-season warming affect leaf senescence over two growing seasons in three tundra species—Artemisia norvegica, Poa alpina, and Trisetum spicatum—collected from the same latitude in northern Canada. Senescence was quantified by digitally measuring leaf colour change in the final weeks of the experimental season. Flowering in one species allowed comparison between foliar and reproductive responses. Photoperiod and temperature both influenced senescence but in opposing ways. All species senesced faster under cooler temperatures, whereas shortening daylength slowed senescence compared to constant daylength. In Poa alpina, reproductive individuals showed faster leaf senescence, and panicle senescence was temperature sensitive. Late-season phenology exhibits species-specific sensitivity to photoperiod and temperature, highlighting variation rarely testable in the field. This could drive increasingly divergent seasonality among tundra plants as climate changes. Foliar and reproductive tissues may respond independently, showing that well-documented early-season responses to northern warming do not always predict late-season phenology.
Dataset DOI: 10.5061/dryad.m905qfvbm
Description of the data and file structure
Files and variables
File: Seed_Registry_Readable.csv
Description: Information that was collected for each seed envelope among all species.
Variables
- Species: Species identifier used during seed collection. These correspond to the scientific names found in the "Scientific_name" column. Note that here "POA" refers to Poa arctica, not Poa alpina as elsewhere.
- Elevation: Elevation in mean metres above sea-level for the collected seeds.
- Plot: Nearest experimental plot, where relevant. If not near any plot, then written as "NA".
- Date: Date of seed collection in 2021. Note that some dates are missing and thus written as "NA".
- Scientific_name: The scientific name of the species collected to best taxonomic resolution. This includes Astragalus alpinus, Artemisia norvegica, Astragalus umbellatus, Draba sp., Polemonium sp., Lupinus arcticus, Oxytropis sp., Primula frigida, Poa arctica, Anemone parviflora, Poa alpina, and Trisetum spicatum.
File: ChamberAnalysisCleaned.R
Description: Code for all analyses.
File: allmeas.csv
Description: All leaf senescence measurements.
Variables
- Species: One of the three species in the experiment. Either "AN", "TS", or "POA" for Artemisia norvegica, Trisetum spicatum, or Poa alpina, respectively.
- ID: Unique individual identifier.
- Week: Experimental measurement week. Here a nonsense variable as all flowers were measured at the same time.
- Areatype: Nonsense variable denoting how area was measured (always 1).
- Area: Total leaf area in square pixels (pixels2).
- Senesced: Proportion of leaf area that is yellow or red (senesced).
- yArea: Senesced (yellow or red) leaf area in square pixels (pixels2).
File: allmeas_flo.csv
Description: All panicle senescence measurements.
Variables
- Species: One of the three species in the experiment. Here always "poa" for Poa alpina.
- ID: Unique individual identifier.
- Week: Experimental measurement week. Here a nonsense variable as all flowers were measured at the same time.
- Areatype: Nonsense variable denoting how area was measured (always 1).
- Area: Total leaf area in square pixels (pixels2).
- Senesced: Proportion of leaf area that is yellow or red (senesced).
- yArea: Senesced (yellow or red) leaf area in square pixels (pixels2).
File: Chamber_conditions.xlsx
Description: A table of experimental chamber conditions.
Variables
- Week: Week of the experimental season.
- Temperature_Day: Temperature during the day (full light).
- Temperature_Night: Temperature during the night (zero light).
- Daylength: Length of the full-light + dawn/dusk segment of each 24 period.
File: ChamberModelComparisonTable.xlsx
Description: A table comparing the AIC values for different model structures.
Variables
- Model: The model in question.
- Group: Which specific group the model was being tested on.
- AIC: The AIC value of the given model.
- ΔAIC: The change in AIC value relative to the lowest AIC value obtained within each group.
- Notes: Notes that would otherwise affect model interpretation and suitability.
File: chambers_assignments_all_readable.csv
Description: Table denoting the assignments of all individuals to all chambers and treatments.
Variables
- Species: One of the three species in the experiment. Either "AN", "TS", or "PA" for Artemisia norvegica, Trisetum spicatum, or Poa alpina, respectively.
- Family: A number denoting which maternal line an individual descends from.
- ID: The unique identifier of the individual.
- Chamber: The chamber the individual was assigned to.
- Treatment: The treatment the individual was assigned to.
- Week: The experiment week, where week 1 & 2 are the 9th and 12th weeks of the first experimental replicated, and week 3 & 4 are the 9th and 12th weeks of the second experimental replicate.
Code/software
R
Microsoft Excel
Access information
Other publicly accessible locations of the data:
- N/A
Data was derived from the following sources:
- N/A
Data were collected from a controlled growth-chamber experiment designed to isolate the effects of late-season photoperiod and temperature on tundra plant senescence. Seeds of three alpine tundra species (Artemisia norvegica, Poa alpina, and Trisetum spicatum) were collected from a single high-latitude population on the Kluane Plateau, Yukon, Canada. Plants were germinated and grown under standardized simulated summer and winter conditions to synchronize developmental stage prior to treatment application.
Individuals were then exposed to a fully factorial set of late-season growth-chamber treatments manipulating daylength (constant vs. shortening) and temperature (constant vs. cooling) over two simulated growing seasons. Treatments reproduced natural and experimentally extended summer conditions to decouple thermal and photoperiodic cues. Leaf senescence progression was quantified using high-resolution digital image analysis of foliage colour change at two time points during the final weeks of each growing season, with additional reproductive tissue measurements for Poa alpina. Mixed-effects modeling was used to evaluate time-dependent and species-specific responses to photoperiod and temperature.