1. To predict shrub responses under climate change in tundra, we need to understand how thermal conditions and herbivory contribute to growth. We hypothesise that shrub growth increases with thermal conditions and precipitation, but that this increase is counteracted by insect herbivory, and that these climate-insect herbivory relationships are modified by both browsing and plant age. 2. We use empirical dynamic modelling (EDM) to analyse a 20-year time series on willow (Salix phylicifolia) shoot growth, growing degree days, summer precipitation and herbivory from an experiment at forest-tundra ecotone. The experiment includes manipulations of avian and mammal browsing (fences) and ramet age (pruning to rejuvenate willows). 3. Negative effects of insect herbivory on willow shoot growth were intensified during warmer years, whereas increasing precipitation led to reduced effects. Moreover, the effect of insect herbivores on shoot growth varied with ramet age and vertebrate browsing: Younger ramets generally experienced less negative insect herbivore effects, whereas Ptarmigan browsing was associated with more positive temperature effect on shoot growth, and reindeer browsing with more negative effects of insect herbivory and precipitation. 4. Synthesis. Our findings show that the negative effects of insect herbivory on shoot growth likely intensify under warmer thermal conditions, but that increasing precipitation can counteract these effects. Moreover, changing thermal conditions, precipitation and vertebrate browsers all have predictable, albeit complex and nonlinear, effects on shrub growth, highlighting the importance of long-term experimental data and flexible analytical methods such as EDM for characterizing climate and community interactions in artic systems.

The study area is located in NW Finnish Lapland at a transitional area between boreal mountain birch forests and treeless oroarctic and arctic tundra heaths at higher elevations (Virtanen et al., 2016). The study site lies at an elevation of 550–570 m a.s.l. (68.92° N, 20.97° E). In total, 40 willow genets in two forest-tundra ecotone sites were selected for the experiment in 1997. In these two sites lying c. 800 m apart, the genets were selected within areas of c. 50 × 200 m and 50 × 250 m. Genets were 30–50 cm tall and consisted of 10–25 ramets each. All genets were initially accessible to both reindeer and ptarmigan. The genets were then assigned into blocks of four genets. In 1997–1998, two random genets within each block were rejuvenated by pruning ramets five cm above the ground level. The pruned genets produced sprouts that initially grew vigorously for 1–3 years and within a few years reached the size of unpruned genets. Thus, each block consisted of two old genets and two rejuvenated genets. Within each block one of the old and rejuvenated genets were protected by reindeer fences (2×2 m area). In 2000, reindeer fences were replaced by ptarmigan proof fences in half of the blocks, selected at random. Final treatments therefore included rejuvination ("J"), reindeer browsing ("R"), and ptarmagin browsing ("P").

Data are in .csv format, with decimal points indicated by ".". "GDD5MJJ" shows summed daily mean temperature in excess of + 5 °C in May–July. "PJune" and "PJuly" show summed annual  precipitation in June and July, in mm. Lastly, columns "s1.cm" through "s11.cm" show shoot lengths -- the mean taken across these columns was the primary response variable in our paper. Finally, the ".R" file includes the source code used to analyse these data.