Title: Svalbard reindeer winter diets: long-term dietary shifts to graminoids in response to a changing climate.
Authors: Tamara A. Hiltunen, Audun Stien, Maria Väisänen, Erik Ropstad4, Jouni O. Aspi, and Jeffery M. Welker

README file to provide key information for understanding and reuse of our data:

The data in the file consist of the following variables (in columns):
Year: 			The year in which the sample was collected. If the sample was collected in 1995 this represented the Winter of 1994/1995.
Sample.Name:		The sample identity that was used to identify the samples. It is composed of the individual reindeers ID number (Ear Tag ID) and the data the sample was collected. 
			For example R42_08.05.95 is reindeer R42 and the sample was collected on the 8th May 1995.
ID: 			Ear tags and neck marker straps on each reindeer were used to record the individual’s identity, 
lweight: 		The live body mass (kg) of the captured individual reindeer on the date of sampling.
pdprog: 		The pregnancy status of the captured individual reindeer on the date of sampling as determined by the presence of progesterone in the blood samples.(0 = Not pregnant; 1 = Pregnant).
d15N			Svalbard reindeer serum d15N values (‰).
d13C.un			Svalbard reindeer serum d13C values (‰) that have NOT been corrected for theanthropogenic increase in CO2 and decrease in d13C.
Schubert.cf		The Schubert correction factor (‰). Anthropogenic inputs of C from fossil fuel burning has resulted in measurable decreases in atmospheric d13C known as the “Suess effect”. 
			In addition, the discrimination of d13C during photosynthesis in C3 plants has been found to increase in response to increasing concentrations of atmospheric CO2 and decreasing atmospheric d13CCO2. 
			We used the equation described by Schubert & Jahren (2012) and annual pCO2 (Tans & Keeling, 2022) to determine the correction factor to correct each years samples d13C values to expected 2012 levels.
d13C			Svalbard reindeer serum d13C values (‰) that have been corrected for the anthropogenic increase in CO2 and decrease in d13C using the Shubert correction factor.
C:N			The ratio of carbon to nitrogen.
ROS.mm			ROS (mm), measured as the sum of precipitation when the average daily temperature was = 1°C during the winter (1st November to the 31st March), as a proxy for ground ice (c/o Peeters et al., 2019). 
July.Temp.Average	July average air temperatures (°C) at Svalbard Airport, obtained from the Norwegian Meteorological Institute, as a proxy for summer plant biomass production 
population		Population density in the Reindalen valley system the previous summer to account for density-dependent competitive effects (c/o Albon et al., 2017)


How the data is used in the various analyses:

Diet proportion modelling (simmr):	Data_file:		Reindeer_data_schubert.xlsx		
					Mix: 			Year, d15N, d13C (Mix = Sheet 1)
					Source: 		Source, d13C, d15N, SDd13C, SDd15N
								(Sources_1 [Model W_13_19 = Sheet 2]; Sources_2 [Model WS_13 = Sheet 3]; Sources_3 [Model WS_13_19 = Sheet 4]; Sources_4 [Model WS_9_13_19 = Sheet 5])
					TDF: 			Source, d13C, d15N, d13C_sd, d15N_sd (TDFs = Sheet 6)
								Imputed in SIDER d13C = 4.51±1.94‰ and d15N = 3.36±1.24‰.
					Rcode:	 		R_Code-simmr_SvalbardReindeer.R

Isotopic Niche modelling (SIBER):	Data_file:		Reindeer_data_schubert_correction.csv
					Data:			d13C, d15N, Year(Group), additional column: Community = 1
					Rcode:			R_Code-SIBER_SvalbardReindeer.R

Linear Mixed Effect Models:		Data_file:		Reindeer_data_schubert_correction.csv
					Data d13C models:	Year, ID, d13C, lweight, Ros.mm (ln transformed), July.Average.Temperature, population (ln transformed) 
					Data d15N models:	Year, ID, d13C, lweight, pdprog, Ros.mm, July.Average.Temperature, population
					Rcode:			R_Code-GLMMs_SvalbardReindeer.R