Climate Windows

Biological Data

The data used in this analysis are listed below.

Helianthella quinquenervis (HEQU)

Iler, A. M., Compagnoni, A., Inouye, D. W., Williams, J. L., CaraDonna, P. J., Anderson, A., & Miller, T. E. (2019). Reproductive losses due to climate change‐induced earlier flowering are not the primary threat to plant population viability in a perennial herb. Journal of Ecology.
Article
Dryad

Cryptantha flava (CRFL)

Salguero-Gómez, R., Kempenich, H., Forseth, I. N., & Casper, B. B. (2014). Long‐term individual‐level population dynamics of a native desert chamaephyte: Ecological Archives E095‐050. Ecology, 95(2), 577-577.
Article
Figshare

Cylindriopuntia imbricata (OPIM)

Provided by Tom Miller. the data spans from 2004 to 2018. Details on data gathering etc can be found in the following articles
Eldred & Miller, 2016
Ohm & Miller, 2014

Frasera speciosa (FRSP)

Provided by David Inouye. See the following websites for more information.
The RMBL Phenology project
Open Science Framework

Climate data

Data on the climate used for this project is retrieved from three sources:

For HEQU: Data from PRISM For CRFL, and FRSP: Data from NOAA is retrieved using the R package rnoaa
For OPIM: Data from the SEV-LTER was used

In case data on specific dates were not available, the values were imputed using 1 or 2 other nearest stations. Any remaining missing data (in case the data was also missing from the other stations) is imputed using the same method as the Climwin's "method1"

Columns	Explanation
Year	year of measurement (for monthly data)
Month	month of measurement (can either be M or MM; for monthly data)
date	date of measurement in "dd\mm\yyyy" (for daily data)
Climate values	Each climate driver has its own column

Metadata

HEQU

Metadata for the HEQU dataset

Columns	Explanation
population	There were three distinct study populations. low: Horse Ranch Park (2703 m a.s.l); mid: Maxfield Meadow, at Rocky Mountain Biological Lab (2886 m a.s.l); high: Virginia Basin (3407 m a.s.l)*
plantID	Unique plant id for each individual (combination of population and plant.ID found in original data)
year	Year of measurements at time T*
month	Month of measurements at time T
day	Day of measurements at time T
presentT	Presence at time T (0/1)*
presentT1	Presence at time T+1 (0/1)*
survival	Survival from year T to year T+1 (0/1)*
seedling	Is individual seedling in year T (0/1)*
sizeT	The number of clumps of leaves per plant at time T*
sizeT1	The number of clumps of leaves per plant at time T+1*
pflowerT	The probability of flowering (producing at least one stalk) at time T*
pflowerT	The probability of flowering (producing at least one stalk) at time T+1*
abort.stalks	Number of aborted flower stalks per plant at time T**
fertilityT	Total number of flowering stalks in which the plant invested at time T*
fertilityT1	Total number of flowering stalks in which the plant invested at time T+1*

* Same as found in the original data; For details see the original README

CRFL

Metadata for the CRFL dataset

Columns	Explanation
Treatment	Treatment applied to the permanent plot. C: Control; D1: Rainout shelter in summer 1998; D2: Rainout shelter in summer 1999*
Block	Location where the treatments were assigned in groups of three plots*
Plot	Plot number. Permanent plots of 5 × 5 m²*
Quadrat	Each permanent plot is conformed by 13 samplable 1 × 1 m² quadrats. Note that in plots 1–9 all quadrats were sampled, but in plots 10–18 only a random subset were sampled.*
plantID	Individual unique identifier, a concatenation of the values for variables "Treatment", "Block", "Plot", "Quadrat", "X" and "Y", separated by "." *
year	Year of census (at time T)*
month	Month of census (at time T)
day	Day of census (at time T)
survival	survival from time T to time T1
sizeT	The total number of rosettes of individual of Cryptantha flava at time T*
sizeT1	The total number of rosettes of individual of Cryptantha flava at time T1*
age	Age of individual of Cryptantha flava. Individuals with unknown age are set to 999*
pflowerT	Did the individual flower
fertilityT	Number of flowering rosettes of individual of Cryptantha flava*
Shrub	Closest shrub to the target individual of Cryptantha flava.*
Compass	Compass direction from the center of the closest shrub to the target individual of Cryptantha flava*
Distance	Closest distance from the target individual of Cryptantha flava to the edge of the shrub. Negative values mean the individual of Cryptantha flava is under the shrub. Positive values mean the individual of Cryptantha flava is not under the shrub.*

* Same as found in the original data. For details see the original README

OPIM

Metadata for the OPIM dataset

Columns	Explanation
Plot	Plot or spatial block*
TagID	Unique identifier for each individual*
Transplant	0 or 1. If 1, the individual was transplanted as part of an old experiment*
year	The calendar year*
month	The calendar month of census
day	The day of census
Height_t	Height of the plant in year t (in cm)*
Width_t	Maximum width of the plant in year t (in cm)*
Perp_t	The width perpendicular to the maximum width in year t (in cm)*
sizeT	The volume of the plant in year t, calculated as a cone, using Height_t, Width_t and Perp_t
Goodbuds_t	The number of viable flowerbuds on a plant in year t
fertilityT	The total number of flowerbuds (viable + inviable) in year t
ABFlowerbuds_t	The number of aborted/inviable flowerbuds in year t
Recruit	0 or 1. Estimated if this is a 0 year recruit in year t**
Newplant	0 or 1. Is this a new plant in the census
Survival_t1	Did the individual survive from year t to year t+1
Height_t1	Height of the plant in year t+1 (in cm)*
Width_t1	Maximum width of the plant in year t+1 (in cm)*
Perp_t1	The width perpendicular to the maximum width in year t+1 (in cm)*
sizeT1	The volume of the plant in year t+1, calculated as a cone, using Height_t1, Width_t1 and Perp_t1
NS_t	The number of new stem segments in year t (a measure of vegetative growth)*
Antcount_t	The number of ants counted in 30 seconds
Antsp_t	The species identity of the ants

* Same as found in the original data
** Personal note from Tom - Not super reliable, very difficult to distinguish recruits from small older plants

FRSP

I used the reformatting script to reformat the original data sheets into a format usuable for this study

Columns	Explanation
year	calendar year of time T
month	month when size was measured at time T. Because I work with a fixed reference date set to 7
day	day when size was measured at time T. Because I work with a fixed reference date set to 15
plantID	Unique identifier of individual.
sizeT	Number of leaves in basal rosettes at time T.
recruitT	Is the individual a new recruit yes (1) or no (0)
sizeT1	Number of leaves in basal rosettes at time T + 1.
survival	if the individual survived from time T to T + 1 (1) or not (0)
pFlowerT1	Did the individual flower