Milner et al. 2016 - Experimental icing affects growth, mortality and flowering in a high Arctic dwarf shrub, Ecol.Evol. DOI - 10.1002/ece3.2023 Raw data from Cassiope icing experiment in high Arctic Svalbard, 2014. Simulated extreme mid-winter rain-on-snow icing event. Experimental design: 2 treatments (control / ice) x 6 plots x 5 replicate Cassiope samples. Two data files 1)exp_icing_dat Data recorded for each Cassiope sample (n=60) are: n.2014flws: number of this year's flowers (2014) - growing at terminal ends, peduncle reddish brown, nearly always have flower head n.allflws: count of all flower present, including grey peduncles with flower heads missing tot.dead.shoots: total number of dead shoots for which tip is still present (i.e. excluding v.old / broken off shoots) tot.n.shoots: count of all shoots present, dead + green n.2014shoots: number of new shoots developed during 2014 (.e. post-treatment) n.damaged.shoots: number of tips showing damage consistent with trauma during 2014 (green but with abnormalities e.g. clusters of multiple new lateral shoots) browning: (categorical) - some of current year's shoots appear dead yet current yr's flowers present (1); no current year's flowers on brown shoots - i.e. dead (0) inc.14: annual growth increment for year 2014 (post-treatment), averaged across shoots from an individual sample (uniquely identified by UID) rel.inc14: inc.14 expressed relative to mean increment for pre-treatment years 2011-2013 at individual sample level (see AGI_Icing) NA: missing value 2) AGI_Icing Cassiope annual growth increment measurements (mm) for pre-treatment years (2011-2013; phase 0) and post-icing treatment (2014; phase 1) for 3 samples per plot per treatment. Relative growth increment (rel.inc) = increment / mean[increment 2011-2013] for each individual shoot (identified uniquely by UID + shoot.rep)