This is the accepted version of the article: Wang, Yafeng; Case, Bradley; Lu, Xiaoming; [et al.]. «Fire facilitates warming- induced upward shifts of alpine treelines by altering interspecific interactions». Trees, structure and function, Vol. 33 (March 2019), p. 1051-1061. DOI 10.1007/s00468-019-01841-6 This version is avaible at https://ddd.uab.cat/record/218326 under the terms of the license 1 Article 2 3 Fire facilitates warming-induced upward shifts of alpine treelines by 4 altering interspecific interactions 5 6 Running head: Fire facilitates upward treeline shifts 7 Yafeng Wang1, 2, Bradley Case3, Xiaoming Lu1, Aaron M. Ellison4, Josep Peñuelas 5, 6, Haifeng 8 Zhu1, 2, 7, Eryuan Liang1, 7*, J. Julio Camarero8 9 1Key 10 Laboratory of Alpine Ecology, Institute of Tibetan Plateau Research, Chinese Academy of 11 Sciences, Beijing 100101, China 12 2College of Biology and the Environment, Nanjing Forestry University, Nanjing 210037, China 3School 13 of Science, Auckland University of Technology, Private Bag 92006, Auckland, 1142, 14 New Zealand 4Harvard 15 University, Harvard Forest, 324 North Main Street, Petersham, MA 01366, USA. 16 5CSIC, Global Ecology Unit CREAF-CSIC-UAB, Bellaterra, E-08193 Catalonia, Spain 6CREAF, 17 Cerdanyola del Vallès, E-08193 Catalonia, Spain 7CAS 18 centre for Excellence in Tibetan Plateau Earth Sciences, Beijing 100101, China 8Instituto 19 Pirenaico de Ecología (IPE, CSIC), Avda. Montañana 1005, 50080 Zaragoza, Spain 20 21 22 *Corresponding author: 23 Eryuan Liang 24 Email: liangey@itpcas.ac.cn 25 Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100101, China. 26 Phone: +86 01084097069 / Fax: +86 01084097069 27 28 1 29 Abstract 30 Key message Moderate-severity fire disturbances can accelerate upslope shifts of alpine treelines 31 by reducing interspecific interactions, providing additional evidence for the species interaction 32 mechanism in controlling treeline dynamics. 33 34 Abstract Biotic interactions between trees and other plants may modulate the responses of alpine 35 treelines to climate. Moderate disturbances could therefore accelerate upward shifts of alpine 36 treelines as the climate warms by reducing the coverage of competitor plants and resetting 37 interspecific interactions. Larch (Larix potaninii var. macrocarpa) treelines disturbed by fire on 38 the southeastern Tibetan Plateau are good locales for testing this hypothesis. We characterized 39 treelines in five large rectangular plots spanning undisturbed and fire-disturbed fir (Abies georgei 40 var. smithii) and larch treelines. The fires in the 1960s caused gaps in the reconstructed age 41 structures of the larches during the 1970s but did not lead to downslope shifts in treeline position. 42 Recruitment has instead increased since the 1980s within the disturbed larch treelines, with 43 treelines shifting upward by 11-44 m. In contrast, the undisturbed larch and fir treeline positions 44 remained mostly unchanged. We conclude that upslope shifts of alpine treelines are likely a 45 consequence of climatic warming, but fire disturbances can accelerate these dynamics by altering 46 interspecific interactions. 47 48 Keywords Climate change · dendroecology · fire disturbance · interspecific competition · tree 49 recruitment · treeline shift. 50 51 2 52 Introduction 53 The structure and position of alpine treelines are considered to be primarily thermally controlled 54 (Körner 2012). In support of this hypothesis, global climatic warming has induced upward treeline 55 encroachment and tree densification at some treeline ecotones (Camarero and Gutiérrez 2004; 56 Chapin et al. 2005; Liang et al. 2011; Liu and Yin 2013; Greenwood et al. 2014; Liang et al. 2016; 57 Wang et al. 2016a). Many treelines globally, however, are relatively stable despite evidence of 58 regional warming (Harsch et al. 2009). Different responses of regional treelines to warming result 59 from variability of the local processes controlling population dynamics at particular treeline 60 locations (Malanson et al. 2007; Gou et al. 2012; Kirdyanov et al. 2012; Case and Duncan 2014; 61 Lyu et al. 2016, Camarero et al. 2017; Wang et al. 2017; Vitali et al., 2019). For example, regional 62 warming promoted an upward shift of undisturbed spruce and fir treelines on the Tibetan Plateau, 63 but their rates of upslope migration were controlled largely by interspecific interactions (Liang et 64 al. 2016). We would therefore expect that upward shifts in treelines could be accelerated or 65 decelerated if competition or facilitation were reduced by local disturbances. 66 As an important agent of disturbance (Swetnam and Betancourt 1990; Brown et al. 1999; 67 Turner et al. 2016), fire can have a wide variety of impacts depending on the severity of the fire 68 and the dynamics of the treeline communities involved (Stueve et al. 2009; Cansler et al. 2018). 69 For example, highly severe fires can destroy vast areas of contiguous subalpine forests and 70 depress alpine treelines, thereby overriding any positive effect of climatic warming on the 71 dynamics of alpine treelines (Hemp 2005; Brown 2010). The roles played by fires of low to 72 moderate severity in mediating alpine-treeline dynamics, however, are less clear, although some 73 evidence suggests that moderately severe fires facilitate tree recruitment into arctic treeless areas 74 by creating favorable microsites for regeneration where competitive plants have been removed 75 (Certini 1993; Landhausser and Wein 1993; Lloyd and Fastie 2003). Few studies, to date, have 76 determined how fire disturbances modulate the responses of alpine treelines to climate, even 77 though fire-disturbed alpine treelines are found in several mountainous areas (Peñuelas and 78 Boada, 2003; Baker and Moseley 2007; Stueve et al. 2009; Aakala et al. 2014). 3 79 The Baima Snow Mountains on the southeastern Tibetan Plateau host larch (Larix potaninii 80 var. macrocarpa) and fir (Abies georgei var. smithii) treelines from 4300 to 4430 m a.s.l. 81 Competition from dense alpine shrublands in these mountains near and above the treeline slows 82 the upward movement of undisturbed treelines induced by warming (Liang et al. 2016). 83 Photographs of landscapes have shown that historical fires have occurred at some treeline 84 ecotones in the mountains (Baker and Moseley 2007). The presence of undisturbed and fire- 85 disturbed treelines in these mountains provides a rare opportunity to investigate the role of fire 86 disturbances in modulating the responses of treelines to climate by relaxing the competition 87 between alpine plants and tree seedlings. 88 In this study, we assessed and further identified the magnitude and drivers of treeline changes 89 using annually resolved tree-ring records and inferred the mechanisms behind interspecific 90 interactions. Specifically, we 1) reconstructed the spatiotemporal patterns of L. potaninii and A. 91 georgei treelines in the Baima Snow Mountains using dendroecological methods, 2) dated recent 92 fires by cross-dating trees killed by fire and 3) tested the hypothesis that fire disturbance facilitates 93 upward shifts of alpine treelines induced by warming by reducing the coverage of plants near or 94 above the treeline. If this hypothesis was confirmed, climate warming would further contribute to 95 upward treeline advances by enhancing tree growth and recruitment but also by increasing fire 96 incidence and altering plant-plant interactions exerted by alpine vegetation on tree recruits. 97 98 Materials and methods 99 Study region 100 The Baima Snow Mountains (27°25′-28°36′N, 98°47′-99°21′E) are in the central Hengduan 101 Mountains, southeastern Tibetan Plateau (Fig. 1). Based on climatic data from the Deqin weather 102 station near the study site (28°23′N, 99°01′E; 4293 m a.s.l.), July (mean monthly air temperature 103 of 7.1 °C) and January (-9.1 °C) are the warmest and coldest months, respectively (Li and Zhang 104 2010). The mean annual precipitation for 1981-2009 was 807 mm, about 70% of which fell during 105 the monsoon season, i.e. from June to August (Li and Zhang 2010). Maximum wind speed varies 4 106 from 3-5 m s-1 throughout the year. The soil is covered by snow (depth ca. 50-100 cm) mainly 107 during winter. Another meteorological station in nearby Deqin county (28°29′N, 99°55′E; 3485 m 108 a.s.l.), about 30 km from the study area, recorded an increase of 2.4 ˚C in summer mean 109 temperatures during 1954-2013 (r = 0.75, P < 0.001), whereas annual or summer precipitation did 110 not show a significant trend during the same period. Temperature reconstructions for the 111 southeastern Tibetan Plateau based on tree rings have also identified significant warming trends in 112 the region during the last 200 years (Zhu et al. 2011). 113 114 Research plots and sampling design 115 Conifer treeline plots were selected in the central Baima Snow Mountains. We defined the treeline 116 and forest limits as the highest elevations where trees were at least 2 m tall and where tree cover 117 was at least 30%, respectively (Holtmeier 2009). The treeline ecotone encompasses both 118 boundaries. The dominant tree and shrub species in the central Baima Snow Mountains include 119 larch (L. potaninii Batal. var. macrocarpa Cheng et Y. W. Law), fir (Abies georgei var. smithii 120 (Viguie et Gaussen) Cheng et L. K. Fu) and a dwarf scrubland-heath complex consisting of several 121 Rhododendron species (e.g. R. phaeochrysum Balf. et W. W. Smith, R. rupicolum W. W. Smith var. 122 chryseum (Balf. Et Ward) Philipson et Philipson, R. nivale Hook. var. boreale Philipson et 123 Philipson, R. tapertiforme Balf. et Ward and R. yungningense Balf.). Alpine meadows are the 124 dominant vegetation type above 4500 m. 125 Larch and fir are the main treeline species in the Baima Snow Mountains. Forests of these 126 species grow along an altitudinal gradient from 3500 m to 4300 m and from 3500 to 4200 m a.s.l., 127 respectively. The positions of the larch and fir treelines vary from 4300 to 4417 m and from 4200 128 to 4450 m, respectively, due to the variable topography and landforms (Table 1). Wind-flagging 129 trees or krummholzs were not observed at the larch and fir treeline. 130 Based on a meta-analysis using a global dataset of 166 treeline sites, there were no clear 131 differences among plant families in their probability of treelie advance (Harsch et al. 2009). On 132 the Tibetan Plateau, we did not find species-related treeline shift rates of four species (Picea 5 133 crassifolia, Picea likiangensis var. balfouriana, Abies georgei var. smithii, and Abies georgei) in 134 the last 100 years (Liang et al. 2016). In the central Himalayas, broadleaf and conifer tree species 135 (Betula utilis and Abies spectabilis) also did not exhibit specific differences in their shift rates of 136 alpine treelines (Sigdel et al. 2018). 137 A number of treelines on the Tibetan Plateau have been rarely disturbed by local human 138 activities (Liang et al., 2011, 2016; Gou et al., 2012; Lyu et al., 2016). Based on weekly 139 xylogenesis monitoring, minimum air temperature was the dominant climatic variable associated 140 with xylem growth at such humid and cold treeline (Rossi et al. 2016; Li et al. 2017). Moreover, 141 biotic competition can affect tree growth and regeneration, and moderate their responses to 142 climate change (Wang et al. 2016a; Liu et al. 2018). Larch seeds could be dispersed upslope by as 143 much as 84–140 m in elevation from parent trees based on the occurrence of larch seedlings above 144 the current treeline (Table S1). As shown by a study on the southeastern Tibetan Plateau, fir 145 seedlings were found 40 m higher above its treeline, implying that seed dispersal was not a 146 constraint for treeline shift (Shen et al. 2014). 147 Some larch treelines in the study area (Fig. 1) have been affected previously by human-caused 148 fire, as indicated by the presence of dead trees and fire-scarred living trees (see more details in 149 Baker and Moseley 2007). Trees killed by fire still had remnants of bark and sapwood on their 150 stems, whereas scarred living trees had fire scars (“cat-faces”) at the stem base (Baker and 151 Moseley 2007). No signs of fires, however, were found across the fir treeline ecotone during field 152 observations from May 2013 to October 2014. 153 Five treeline sites (one fir and four larch sites) on different topographical conditions were 154 selected in the Baima Snow Mountains (Table 1). Rectangular plots with size 30 m× 200 m (2 155 plots), 30 m × 270 m, 30 m × 280 m and 30 m × 500 m were established on topographically 156 uniform areas across treeline ecotones and included the current treelines and forest limits. These 157 treeline plots were surveyed following the methods of previous treeline studies on the Tibetan 158 Plateau (Liang et al. 2011, 2016; Wang et al. 2016b). The longest side (y-axis) of each plot was set 159 parallel to the steepest slope (i.e. following the altitudinal gradient), and the origin (x, y) = (0, 0) 6 160 was located in its bottom left corner. The altitudes of the lower and upper parts of the plots were 161 determined using a GPS calibrated with a barometric altimeter (Cook and Kairiukstis 1990). The 162 location of each visible tree individual within the plot was mapped (± 0.1 m) using x-y coordinates 163 relative to the plot origin. Other biometric parameters measured and recorded for each tree 164 included: DBH (diameter at breast height, measured at 1.3 m aboveground), tree height, and 165 diameter of the tree canopy horizontally projected along the x and y axes. Tree height was 166 determined with a measuring stick for trees ≤ 2 m (± 0.01 m), or a clinometer for trees > 2 m (± 167 0.1 m) (Duncan, 1989). 168 Samples were taken from all recorded trees using an increment borer. Ages of trees > 2.5 m 169 were determined using standard dendrochronological methods after air drying cores, sanding them 170 with successively finer sandpapers, and visually cross-dating them under a stereomicroscope 171 (Liang et al. 2011). If the pith was absent in a core, a pith geometric locator was employed to 172 estimate the distance missing up to the center of the stem (Duncan 1989). The mean series 173 intercorrelation varied from 0.58 to 0.62, indicating a robust cross-dating of annual ring-width 174 series in the five study plots (see also the results in Wang et al. 2012). 175 The age structure of the trees was obtained by combining several methods. The ages of 176 seedlings (height ≤ 0.5 m) and juveniles (0.5m < height < 2.5 m) were non-destructively 177 determined in the field by counting the branch whorls along the main stem (Camarero and 178 Gutiérrez 2004). We randomly collected stem disks from 30 seedlings located immediately outside 179 the study plots and also counted the bud scars for these seedlings. The ages of seedlings obtained 180 from counting the internodes (CI) and the rings of stem disks (CR) were significantly correlated 181 (CI = 0.76* CR + 2.42; r = 0.77, P < 0.0001, n=30) and reliable for determining tree age (Aakala 182 et al. 2014), although counting the bud scars underestimated the true seedling age by as much as 183 1-4 years (Figure S1). 184 We determined the age of 150 larch individuals and 30 fir individuals at or near the treelines, 185 respectively, by counting the internodes to obtain more accurate estimates of tree ages in the two 186 height classes (1.3-2.5 and ≥ 2.5 m). The larch and fir trees required averages of 18 ± 2 (n=150) 7 187 and 33 ± 5 (n=30) years to reach 2 m in height, respectively. As suggested by previous treeline 188 studies (Camarero and Gutiérrez 2004; Liang et al. 2011, 2016), we assumed that a seedling 189 required the same time (18 and 33 years for larch and fir treelines, respectively) to reach 2 m 190 during the past and current climatic conditions. The mean DBH for two undisturbed plots and 191 three fire-disturbed plots was 19.7 ± 12.4 (N1, n = 145), 11.2 ± 14.7 (NE3, n = 286) and 8.1 ± 6.5 192 (E1, n = 93), 5.3 ± 7.1 (NE1, n = 99), 6.7 ± 9.7 (NE2, n = 92), respectively. The mean height for 193 two undisturbed plots and three fire-disturbed plots was 8.5 ± 5.5 (N1, n = 167), 2.5 ± 4.6 (NE3, n 194 = 777) and 2.0 ± 2.0 (E1, n = 246), 1.1 ± 1.3 (NE1, n = 458), 1.6 ± 2.1(NE2, n = 270), 195 respectively. 196 Fire-scarred living or dead trees were found in the three larch plots (E1, NE1 and NE2) but not 197 in the NE3 and N1 plots. The alpha-numeric codes correspond to treeline plots located on different 198 slope aspects. For instance, E1 represents one plot on east-facing slope. We estimated the years of 199 fires in the E1, NE1 and NE2 plots by collecting 34 cores 1.3 m above the ground from 17 trees 200 killed by fire. We also collected 30 cores from mature living trees near the undisturbed larch forest 201 limit. The sampled dead trees had retained the main stem bark and sapwood, so the year of a 202 recent fire was identified by cross-dating the tree-ring samples from the living trees at the 203 undisturbed larch treelines with those from the dead trees at the fire-disturbed larch treelines. 204 Treeline position was defined where erect trees had a minimum height of 2 m at the uppermost 205 elevation within each plot (Camarero and Gutiérrez 2004; Liang et al., 2011). The treeline position 206 at 50-year intervals was reconstructed by calculating the years required for trees to reach 2 m in 207 height (Liang et al. 2011). The 4-year error associated with age estimation for trees 2 m tall was 208 negligible for the time scales of 50 years used in the reconstructions of the treeline dynamics 209 reported here. 210 The mean height and canopy cover of vegetation above the treeline were surveyed as described 211 by our previous treeline study (Liang et al. 2016). Along the elevational (y) axis of each treeline 212 plot, six line-transects were established from the upper treeline to 20 m above the treeline. 213 Transects were spaced 6-m apart perpendicular to the slope: that is, along the axis perpendicular to 8 214 the maximum slope (x axis) at 0, 6, 12, 18, 24, and 30 m. Percent vegetation cover was calculated 215 based on the number of contacts between plants and a rod located every meter along the 216 elevational axis, and the height of the vegetation in each treeline plot was calculated as the mean 217 height of the vegetation at each sample point. A total of 126 points were recorded above the 218 treeline for each plot. The thickness index (TI), an ecological indicator of interspecific interaction, 219 is defined as the product of alpine shrub cover above the treeline multiplied by mean shrub height 220 (Liang et al. 2016). Such line-transects were set up in each study site. 221 222 Methods 223 Surviving trees are likely to grow very slowly in the year after crown-damaging fires (Brow and 224 Kaufmann 1999), so we identified the year in which growth rates were suppressed and assumed 225 that was when the historical fire occurred. We used t-tests to compare the mean post-fire tree-ring 226 widths of living and fire-killed trees in the final years (e.g. 1969, 1970, 1971) to determine when 227 growth was suppressed. Very narrow rings in some years usually indicated the presence of 228 disturbance (Liang et al., 2015). Samples from fire-scarred living trees were also used to 229 determine the years of the fires. The ages of 30 randomly selected young trees near or around the 230 fire-scarred trees and probably recruited after the fire was estimated by counting their bud scars. 231 Fire occurrence (binary variable) was related to TI to explore their relationships. 232 Climatic records for the study area are very short and usually available only from the 1950s 233 onwards, so temperature reconstructions based on independent dendrochronological proxies were 234 used to evaluate the long-term impact of climate on changes in recruitment. We used the series for 235 summer mean minimum temperatures derived from a chronology of tree-ring widths for Balfour 236 spruce (Picea likiangensis (Franch.) Pritz. var. balfouriana (Rehd. et Wils.) Hill. ex Slavin) 237 developed on the southeastern Tibetan Plateau (Zhu et al. 2011). The data for reconstructed 238 temperatures indicated a substantial warming trend in the last 200 years (P < 0.01, slope of linear 239 fit: 0.012) (Zhu et al. 2011). 9 240 Age structure of living trees represents the tradeoff between recruitment and mortality through 241 time (Camarero and Gutiérrez 2004; Liang et al. 2011; Wang et al. 2016a). Since recruitment rate 242 of trees is mainly limited by low temperature at undisturbed treelines (Liang et al. 2016; Wang et 243 al. 2016a), it is reasonable to assume that expanding population structure appears under climate 244 warming, and vice-versa. In this study, the recruitment history reconstructed by age structure was 245 consistent with the repeated landscape photography at Larix treelines in the Baima Snow 246 Mountains located within our study region (Baker and Moseley 2007). Therefore, we used the 247 reconstructed age structure at treeline plots to discuss changes in tree recruitment through time. 248 Post-fire tree recruitment data in the three fire-disturbed larch plots were compared to 249 determine the effect of fire disturbance on the pattern of tree recruitment at treeline ecotones. Post- 250 fire recruitments for these plots were then compared with those of the undisturbed larch and fir 251 treeline plots. The two-sample Kolmogorov-Smirnov (K-S) test was used to test if the distribution 252 of recruitment series differed significantly within the three fire-disturbed plots. The K-S test was 253 similarly used to identify significant differences in tree recruitment between fire-disturbed and 254 undisturbed plots. 255 We used a sequential algorithm developed by Rodionov (2004) and described by Elliott (2012) 256 to detect abrupt changes in the reconstructed patterns of tree establishment. This is a data-driven 257 approach in which an a priori hypothesis on the timing of regime shifts is not needed, and 258 sequential t tests are performed on time-series data to detect regime shift changes (Rodionov 259 2004). An abrupt change was identified if the cumulative sum of normalized deviations from the 260 mean of a potentially new pattern of establishment differed significantly from the mean of the 261 previous pattern of establishment episodes (Elliott 2012). A significance level of 0.05 and a 7-year 262 cut length were used to test for shifts in the means of the time series of the recruited trees. The 263 analyses of abrupt changes used the STARS program (Rodionov 2004). Such a method has been 264 used to analyze abrupt changes in the recruitment data at treelines in the Rocky Mountains, USA 265 (Elliott 2012). 266 10 267 Results 268 Fire-induced changes in tree growth were compared to determine fire events at the treelines. The 269 maximum and mean lengths of the series of tree-ring widths were 249 and 132 years, respectively. 270 The series of fire-killed trees was cross-dated with series of living trees, with average inter-series 271 correlation of 0.63 (P < 0.001). The year of the outer ring in the trees killed by fire varied from 272 1969 to 1971. The mean tree-ring widths of the dead trees in 1969, 1970 and 1971 were 273 significantly narrower than those of the surviving trees (P < 0.05 for 1969, 1970 and 1971; n = 30 274 trees). A very narrow ring formed in 1969 was consistently observed in the dead trees (Fig. 2). 275 Fire-scars in the surviving larch trees were also dated from 1969 to 1971. Thus, the three years 276 with narrow tree rings represent the impacts of fire on the individual tree. 277 Age structures and recruitment dynamics vary as functions of climate and fire. The estimated 278 year of the fire in the late 1960s (Fig. 2) coincided with the absence of recruits in the NE2 plots 279 and a much reduced tree establishment in the E1 and NE1 plots during the 1970s (Fig. 3b, d). 280 Young trees recruited since the 1990s accounted for the largest proportion of the treeline 281 populations (Fig. 3) and had an approximately reverse J-shaped age structure. Recruitment pulses 282 in the undisturbed larch N1 plot and fir NE3 plot were characterized by wave-like and rapidly 283 rising recruitment patterns, respectively (Fig. 3e, f). Tree recruitment in the three fire-disturbed 284 larch treelines or two undisturbed treelines did not differ significantly (P > 0.05, K-S tests) 285 between plots (E1 vs. NE1, E1 vs. NE2 or NE1 vs. NE2; NE3 vs. NE1) (Table 2). In contrast, tree 286 recruitment between the fire-disturbed and undisturbed larch/fir plots differed significantly (P < 287 0.05, K-S tests on E1 vs. N1/ NE3, NE1 vs. N1/ NE3 and NE2 vs. N1/ NE3) (Table 2). Regardless 288 of the local site conditions, recruitment has increased in recent decades for the five study plots 289 (Fig. 3b-f) and was positively correlated with reconstructed summer temperatures on the 290 southeastern Tibetan Plateau (Fig. 3a) (plot E1: r = 0.59, n = 8; plot N1: r = 0.57, n = 32; plot 291 NE1: r = 0.56, n = 16; plot NE2: r = 0.44, n = 26; plot NE3: r = 0.71, n = 24; P < 0.05 for all plots 292 except E1; see also Fig. 3). 11 293 Abrupt changes in recruitment were detected within treeline plots. Tree establishment increased 294 abruptly in the 1990s within the three fire-disturbed plots (E1, NE1 and NE2; see Fig. 3b-d). 295 Recruitment also increased abruptly in the 1970s in the undisturbed fir NE3 plot (Fig. 3e) and in 296 the 1980s in the undisturbed larch N1 plot (Fig. 3f). An abrupt change in recruitment in the 1900s 297 was also detected in the N1 plot (Fig. 3f). 298 Based on population age structure in the investigated treeline plots, we reconstructed treeline 299 elevation based on the presence of the tallest trees (≥ 2 m) at a 50-y interval. The treelines in the 300 undisturbed larch (N1) and fir (NE3) plots advanced 0 and 2 m in elevation during the last 50 301 years but advanced 4 and 2 m in elevation during the last 100 years, respectively (Table 3; Fig. 4 302 a, b). The larch treeline plots (E1, NE1, NE2) contained some fire-induced dead trees, but the fire 303 did not cause a downslope shift of the treelines (Table 3; Fig. 4c-e). Furthermore, densification 304 mainly occured above the post-fire treeline. The fire-disturbed larch treeline in the E1, NE1 and 305 NE2 plots advanced 11 m, 44 m and 22 m in elevation in the last 50 years, respectively, and the 306 larch treeline in plot NE2 shifted upward by 22 m in the last 100 years (Table 3; Fig. 4). The fire- 307 disturbed treelines have advanced by 11-44 m since the 1980s (Table 3; Fig. 4). Fire disturbance 308 explained 91.7-100% of the larch-treeline dynamics for the last 50 years (Table 3). In particular, 309 the treeline in the NE2 plot during the last 100 years advanced only after the fire disturbance. 310 Coverage of alpine shrubs above the treeline was related to fire occurrence. Dense 311 Rhododendron shrubs dominated the undisturbed treelines (NE3 and N1), with TIs of 1.64 and 312 1.87, respectively (Table 3). In contrast, shrubs dominated the fire-disturbed treelines, with TIs 313 ranging from 0.23 to 0.67 (Table 3). 314 315 Discussion 316 The warming in the last century is unprecedented over the last 1,000 yrs on the Tibetan Plateau (Liu 317 et al. 2005; Bräuning 2006; Thompson et al. 2006; Zhu et al. 2008). Global warming is expected to 318 have considerable effects on species’ range limit such as alpine treelines, where tree growth is 319 primarily limited by low temperatures (Fang et al. 2009; Liu et al. 2009; Yang et al. 2009; Zhu et al. 12 320 2011). However, climate warming is only one aspect of environmental change that could alter the 321 structure and position of treelines (Carrer and Urbinati 2006; Batllori and Gutiérrez 2008; Liang 322 and Camarero 2018). Apart from climate, other factors such as fire disturbance could buffer or 323 amplify warming impacts on treeline dynamics locally (Aakala et al. 2014). Given a wide 324 occurrence of fire disturbance in subalpine forests (Yao et al. 2017; Waito et al. 2018), it is a 325 challenge to investigate how fire disturbances mediate the responses of treelines to climate warming. 326 We found that treeline dynamics were jointly driven by fire and climatic warming. At least one 327 fire in the late 1960s drastically affected tree recruitment at the treeline. The lack of tree 328 establishment in fire-disturbed larch treelines during the 1970s could be attributed to the adverse 329 effect of a recent fire, given that the favorable warming conditions since the 1950s on the Tibetan 330 Plateau are generally linked to increased tree growth and recruitment (Liu et al. 2009; Yang et al. 331 2009; Liang et al. 2011, 2016; Liu et al. 2011; Zhu et al. 2011; Chen et al. 2016; Wang et al. 2016a). 332 The initial decrease in recruitment during the first 10 years after the fire could have been due to 333 damage to tree crowns and reduced seed production or to the lack of facilitation from short shrubs 334 which could become sparse after fire disturbance. As a control, a larger episode of recruitment in 335 the 1970s was detected at the undisturbed larch and fir treelines, providing indirect evidence for this 336 hypothesis. Recruitment at fire-disturbed sites, however, increased in the 1980s and increased 337 abruptly in the 1990s, suggesting a major acceleration in recruitment relative to the unburned sites, 338 which remained stable. This finding indicates that disturbances play a pivotal role in mediating the 339 responses of treelines to climatic warming (Aakala et al. 2014), as they do in forests (Turner et al. 340 2016). Young larch trees ranging in age from 20 to 30 years were found near or around the fire- 341 scarred trees across the treeline ecotones. This pattern suggests that a post-fire recruitment release 342 took place from the 1980s onwards, which would have allowed for the growth of post-fire recruited 343 seedlings and them reaching heights ≥ 2 m in recent decades. 344 Climatic warming may have a larger impact on the density than the position of undisturbed fir 345 treelines, whereas nearby fire-disturbed treelines have advanced substantially in the last century. A 346 lack of disturbance may thus be responsible for the apparent fir-treeline stability, despite the trends 13 347 of increasing temperature. A previous study indicated that fire was used as a traditional tool to halt 348 the advance of woody species in the Baima Snow Mountains (Baker and Moseley, 2007). Thus, it 349 is likely that the 1969 fire was anthropogenic in origin rather than the more frequent “natural” 350 fires that occur at high latitudes (Yao et al. 2017; Waito et al. 2018). Grazing disturbance also 351 cannot be excluded. Comparative research on the treeline ecotone in other areas of the Tibetan 352 Plateau under human impact has found effects of animal grazing, especially in sites where nearby 353 human settlements exist (e.g. Miehe and Miehe, 2000). Grazing pressure may explain the lack of 354 recruitment before the 1960s at disturbed treelines. However, considering that livestock grazing 355 has always occurred near the disturbed treelines and that grazing intensity tends to increase with 356 the rise of local population (Baker and Moseley, 2007), grazing pressure did not likely reverse the 357 increasing trend in tree recruitment and limit treeline expansion. 358 Historical fire disturbances may have promoted upward shifts of treelines by reducing the 359 competitive pressure from alpine plants. A past fire had an initial negative effect on the 360 recruitment dynamics in the 1970s at the larch treelines but did not depress their elevations. More 361 importantly, trees have encroached and densified in the areas beyond the post-fire treeline 362 elevation, indicating a large upward treeline shift after the fire. The magnitude of advancement of 363 fire-disturbed larch treelines since 1980 explained most of the treeline advance during the last 50 364 and 100 years, suggesting that fire has played an important role in driving the upward movement 365 of treelines. In contrast, undisturbed larch or fir treelines have remained stable, providing further 366 evidence that fire could trigger or even promote upward treeline shifts. Fire may presumably 367 create favorable microsites for successful recruitment by removing competing vegetation and 368 clearing dense shrubs. 369 The TIs calculated for the alpine plant communities near or above the alpine treeline indicated 370 that the intensity of interspecific interactions differed between the undisturbed and fire-disturbed 371 treelines. Tree–shrub competition at undisturbed treelines is likely to limit recruitment of tree 372 seedlings for TIs > 0.8 (Liang et al. 2016). TIs < 0.8 at fire-disturbed treelines represented 373 conditions where low-density shrubs did not prevent tree recruitment above the treeline but 14 374 created environmental conditions suitable for tree establishment (Liang et al. 2016). A previous 375 study found that alpine shrubs (Artemisia tridentata Nutt. ssp. vaseyana (Rydb.) Beetle and 376 Purshia tridentata (Pursh) DC) in the Intermountain West of USA need > 50 years to recover after 377 moderately severe fires (Ziegenhagen and Miller 2009). On the southeastern Tibetan Plateau, 378 Rhododendron shrubs can reach an age of up 400 years (Lu et al. 2015; Bi et al. 2017). In harsh 379 environments at alpine treelines, recovery after fire is slow. Thus, fire may lead to substantial 380 changes in shrub cover, not shrub species composition. Current TIs, therefore, are likely to be 381 much higher than those observed immediately after fire events. Over the past 30 years, post-fire 382 tree-shrub interactions might have changed from low-intensity facilitation to high-intensity 383 facilitation at and beyond the treeline. These results indicated that recent fires may alter species 384 interactions and hence modulate treeline responses to warming. 385 Other studies in boreal forests also showed that moderate-severity fire disturbances can 386 potentially accelerate the northward movement of treelines by positively influencing biotic factors 387 such as seed availability and seedling establishment, supporting our main results. In fact, the 388 ecological benefits of fire disturbances have been recognized (Stephens and Ruth 2005; Carswell 389 2018). However, high-severity fire disturbance would result in the contractions of forest 390 distribution and treeline retreat, thereby posing a major threat to some forest types (Bond and 391 Keeley 2005; Rolstad et al. 2017). In addition, the impacts of high-severity fires on forests might 392 be linked to tree species-specific traits (e.g., regeneration or adaptive strategy). For instance, North 393 American boreal black spruce forest generally takes advantage of fire because it produces semi- 394 serotinous cones that release seeds massively during a fire event and contribute to treeline 395 regeneration and expansion (Auger and Payette 2010). These studies highlighted that a deeper 396 understanding of fire-disturbed treeline dynamics under climate change should consider non- 397 climatic factors. 398 In summary, upward shifts of conifer treelines at our study sites likely were a consequence of 399 climatic warming, but disturbance by fire may have accelerated the rates by reducing competition 400 between tree seedlings and alpine shrubs. The elevation of undisturbed fir and larch treelines has 15 401 remained relatively stable despite climatic warming, presumably due to stronger interspecific 402 competition. Fires of low to moderate severity and other disturbances (e.g. grazing), which relax 403 negative species interactions, could therefore act as triggers to accelerate treeline dynamics and 404 couple these dynamics more strongly to climatic warming. As shown by this study, fire 405 disturbance provides a simplified model to detect how species interactions drive treeline shifts. It 406 provides additional evidence for the species interaction mechanism in controlling treeline 407 formation and its response to climatic change. Further work is expected to quantitatively evaluate 408 the effects of different fire severities on species interactions as well as on treeline shift rates. 409 410 Author contribution statement E.L. designed the research; Y.W., X.L. and E.L. performed the 411 research; All authors analyzed data and wrote the paper. 412 413 Acknowledgments This work was supported by the National Natural Science Foundation of 414 China (41661144040, 41525001), and the International Partnership Program of the Chinese 415 Academy of Sciences (131C11KYSB20160061). AME’s participation in this project was 416 supported by NSF grant DEB 1237491, the Harvard Forest and the Chinese Academy of Sciences 417 (CAS) Presidential International Fellowship Initiative for Visiting Scientists, Grant no. 418 2016VB074. JP’s participation in this project was supported by the European Research Council 419 Synergy grant SyG-2013-610028 IMBALANCE-P. We thank Xuefei Yang, Yingfeng Bi, Shibin 420 Jiao for fieldwork assistance. 421 422 Compliance with ethical standards 423 Conflict of interest We declare that there are no conflicts of interest. 424 425 References 426 Aakala T, Hari P, Dengel S, Nwberry SL, Mizunuma T, Grace J (2014) Prominent stepwise 427 advance of the treeline in north-east Finland. J Ecol 102:1582–1591. 16 428 Auger S, Payette S (2010) Four millennia of woodland structure and dynamics at the Arctic 429 treeline of eastern Canada. Ecology 91:1367–1379. 430 Baker BB, Moseley RK (2007) Advancing treeline and retreating glaciers: Implications for 431 conservation in Yunnan, PR China. Arct Antarc Alp Res 39:200–209. 432 Batllori E, Gutiérrez E (2008) Regional tree line dynamics in response to global change in the 433 Pyrenees. J Ecol 96:1275–1288. 434 Bi Y, Xu J, Yang J, Li Z, Gebrekirstos A, Liang E, Zhang, S, Yang Y, Yang Y, Yang X (2017) Ring 435 ‑widths of the above tree-line shrub Rhododendron reveal the change of minimum winter 436 temperature over the past 211 years in Southwestern China. Clim Dyn 48:3919–3933. 437 Bräuning A (2006) Tree-ring evidence of ‘Little Ice Age’ glacier advances in southern Tibet. 438 Holocene 16:369-380. 439 Bond WJ, Keeley J E (2005) Fire as a global 'herbivore': the ecology and evolution of flammable 440 ecosystems. Trends Ecol Evol 20:387–394. 441 Brown CD (2010) Tree-line dynamics: adding fire to climate change prediction. Arctic 63:488– 442 492. 443 Brown PM, Kaufmann MR, Shepperd WD (1999). Long-term landscape patterns of past fire 444 events in a montane ponderosa pine forest of central Colorado. Landscape Ecol 14:513–532. 445 Camarero JJ, Gutiérrez E (2004) Pace and pattern of recent treeline dynamics: response of 446 ecotones to climatic variability in the Spanish Pyrenees. Clim Change 63:181–200. 447 Camarero JJ, Linares JC, García-Cervigón AI, Batllori E, Martínez I, Gutiérrez E (2017) Back to 448 the future: the responses of alpine treelines to climate warming are constrained by the current 449 ecotone structure. Ecosystems 20:683–700. 450 Cansler CA, McKenzie D, Halpern CB (2018) Fire enhances the complexity of forest structure in 451 alpine treeline ecotones. Ecosphere 9:e02091. 452 Carrer M, Urbinati C (2006) Long‐term change in the sensitivity of tree‐ring growth to climate 453 forcing in Larix decidua. New Phytologist 170:861-872. 454 Carswell C (2018) Tree rings reveal increased fire risk for southwestern US. Nature 554:283–284. 17 455 Case BS, Duncan RP (2014) A novel framework for disentangling the scale-dependent influences 456 of abiotic factors on alpine treeline position. Ecography 37:838–851. 457 Certini G (1993) Effects of fire on properties of forest soils: a review. Oecologia 143:1–10. 458 Chapin FS, Sturm M, Serreze MC, et al. (2005) Role of land-surface changes in arctic summer 459 warming. Science 310:657–660. 460 Chen F, Zhang Y, Shao X, Li M, Yin ZY (2016) A 2000-year temperature reconstruction in the 461 Animaqin Mountains of Tibet Plateau, China. Holocene 26:1904–1913. 462 Cook ER, Kairiukstis L A (1990) Methods of dendrochronology: applications in the environmental 463 sciences. Kluwer, London. 464 Duncan RP (1989) An evaluation of errors in tree age estimates based on increment cores in 465 Kahikatea (Dacrycarpus Dacrydioides). New Zealand Nat Sci 16:31–37. 466 Elliott GP (2012). Extrinsic regemie shift drive abrupt changes in regeneration dynamics at upper 467 treeline in the Rocky Mountains, USA. Ecology 93:1614–1625. 468 Fang K, Gou X, Chen F, Peng J, D’Arrigo R, Wright W, Li M-H. 2009. Response of regional tree- 469 line forests to climate change: evidence from the northeastern Tibetan Plateau. Trees 23:1321– 470 1329. 471 Gou X, Zhang F, Deng Y, Ettl GJ, Yang M, Gao L, Fang K (2012) Patterns and dynamics of tree- 472 line responses to climate change in the eastern Qilian Mountains, northwestern China. 473 Dendrochronologia 30:121–126. 474 Greenwood S, Chen J, Chen C, Jump A (2014) Strong topographic sheltering effects lead to 475 spatailly complex treeline advance and increase forest density in a subtropical mountain region. 476 Glob Chang Biol 20:3756–3766. 477 Harsch MA, Hulme PE, McGlone MS, Duncan RP (2009) Are treelines advancing? A global meta- 478 analysis of treeline response to climate warming. Ecol Lett 12:1040–1049. 479 Hemp A (2005) Climate change-driven forest fires marginalize the impact of ice cap wasting on 480 Kilimanjaro. Glob Chang Biol 11:1013–1023. 481 Holtmeier FK (2009) Mountain timberlines: Ecology, patchiness and dynamics. Springer, Berlin. 18 482 Kirdyanov AV, Hagedorn F, Knorre AA, et al. (2012) 20th century tree-line advance and 483 vegetation changes along an altitudinal transect in the Putorana Mountains, northern Siberia. 484 Boreas 41:56–67. 485 Körner C (2012) Alpine Treelines: Functional ecology of the global high elevation tree limits. 486 Springer, Berlin. 487 Landhausser SM, Wein RW (1993) Postfire vegetation recovery and tree establishment at the 488 arctic treeline: Climate-change-vegetation-response hypotheses. J Ecol 81:665–672. 489 Li W, Zhang Y (2010). Vertical climates in the Hengduan Mountains and their impact on forest 490 distribution. China Meterological Press, Beijing. 491 Li X, Liang E, Gričar J, Rossi S, Čufar K, Ellison AM (2017) Critical minimum temperature limits 492 xylogenesis and maintains treelines on the southeastern Tibetan Plateau. Sci Bull 62:804–812. 493 Liang E, Camarero JJ (2018) Threshold-dependent and non-linear associations between 494 temperature and tree growth at and below the alpine treeline. Trees 32:661–662. 495 Liang E, Leuschner C, Dulamsuren C, Wagner B, Hauck M (2015) Global warming-related tree 496 growth decline and mortality on the north-eastern Tibetan plateau. Climatic Change 134:163- 497 176. 498 Liang E, Wang Y, Eckstein D, Luo T (2011) Little change in the fir tree-line position on the 499 southeastern Tibetan Plateau after 200 years of warming. New Phytol 190:760–769. 500 Liang E, Wang Y, Piao S, Lu X, Camarero JJ, Zhu H, Zhu L, Ellison AM, Ciais P, Peñuelas J 501 (2016) Species interactions slow warming-induced upward shifts of treelines on the Tibetan 502 Plateau. P Natl Acad Sci USA 113:4380–4385. 503 Liu H, Yin Y (2013) Response of forest distribution to past climate change: An insight into future 504 predictions. Chin Sci Bull 58:4426–4436. 505 Liu S, Li X, Rossi S, Wang L, Li W, Liang E, Leavitt SW (2018) Differences in xylogenesis 506 between dominant and suppressed trees. Am J Bot 105:950–956. 507 Liu X, Qin D, Shao X, Chen T, Ren JW (2005) Temperature variations recovered from tree-rings 508 in the middle Qilian Mountain over the last millennium. Sci China Ser D Earth Sci 48:521–529. 19 509 Liu Y, An Z, Linderholm HW, et al (2009) Annual temperatures during the last 2485 years in the 510 mid-eastern Tibetan Plateau inferred from tree rings. Sci China Ser D Earth Sci 52:348–359. 511 Lloyd AH, Fastie CL (2003) Recent changes in treeline forest distribution and structure in interior 512 Alaska. Ecoscience 10:176–185. 513 Lu X, Camarero JJ, Wang Y, Liang E, Eckstein D (2015) Up to 400-year old Rhododendron 514 shrubs on the southeastern Tibetan Plateau: prospects for shrub-based dendrochronology. 515 Boreas 44:760–768. 516 Lyu L, Zhang Q, Deng X, Mäkinen H (2016) Fine-scale distribution of treeline trees and the nurse 517 plant facilitation on the eastern Tibetan Plateau. Ecol Indic 66:251–258. 518 Malanson GP, Butler DR, Fagre DB, et al (2007) Apine treeline of western north America: Linking 519 organism-to-landscape dynamics. Phys Geogr 28:378–396. 520 Miehe G, Miehe S (2000) Comparative high mountain research on the treeline ectone under 521 human impact: Carl Troll' "Asymetrical zonation of the humid vegetation types of the world" of 522 1948 reconsidered. Erdkunde 54:34-50. 523 Peñuelas J, Boada M (2003) A global change-induced biome shift in the Montseny mountains (NE 524 Spain). Glob Chang Biol 9:131–140. 525 Rodionov, S. N. (2004). A sequential algorithm for testing climate regime shifts. Geophys Res Lett 526 31:111–142. 527 Rolstad J, Blanck Y, Storaunet KO (2017) Fire history in a western Fennoscandian boreal forest as 528 influenced by human land use and climate. Ecol Monogr 87:219–245. 529 Rossi S, Anfodillo T, Cufar K, et al (2016). Pattern of xylem phenology in conifers of cold 530 ecosystems at the Northern Hemisphere. Glob Chang Biol 22:3804–3813. 531 Shen W, Zhang L, Liu X, Luo T (2014) Seed-based treeline seedlings are vulnerable to freezing 532 events in the early growing season under a warmer climate: Evidence from a reciprocal 533 transplant experiment in the Sergyemla Mountains, southeast Tibet. Agr Forest Meteorol 534 187:83–92. 20 535 Sigdel SR, Wang Y, Camarero JJ, Zhu H, Liang E, Peñuelas J (2018) Moisture-mediated 536 responsiveness of treeline shifts to global warming in the Himalayas. Glob Chang Biol doi: 537 10.1111/gcb.14428 538 Stephens SL, Ruth LW (2005) Federal forest-fire policy in the United States. Ecol Appl 15:532– 539 542. 540 Stueve KM, Cerney DL, Rochefort RM, Kurth LL (2009) Post-fire establishment patterns at the 541 alpine treeline ecotone: Mountain rainier National Park, Washington, USA. J Veg Sci 20:107– 542 120. 543 Swetnam TW, Betancourt JL (1990) Fire-Southern Oscillation relations in the Southwestern 544 United States. Science 249:1017–1020. 545 Thompson LG, Mosley-Thompson E, Brecher H, Davis M, León B, Les D, Lin P-N, Mashiotta T, 546 Mountain K (2006) Abrupt tropical climate change: past and present. Proc Nat Aca Sci USA 547 103:10536–10543. 548 Turner MG, Whitby TG, Tinker DB, Romme WH (2016) Twenty-four years after the Yellowstone 549 fires: Are postfire lodgepole pine stands converging in structure and function? Ecology 550 97:1260–1273. 551 Vitali A, Garbarino M, Camarero JJ, Malandr F, Toromani E, Spalevic V, Ćurović M, Urbinati C 552 (2019) Pine recolonization dynamics in Mediterranean human-disturbed treeline ecotones. 553 Forest Ecol Manag 435:28-37. 554 Waito J, Girardin MP, Tardif JC, Conciatori F, Bergeron Y, Ali AA (2018) Recent fire activity in 555 the boreal eastern interior of North America is below that of the past 2000 yr. Ecosphere 556 9:e02287. 557 Wang Y, Camarero JJ, Luo T, Liang E (2012) Spatial patterns of Smith fir alpine treelines on the 558 south-eastern Tibetan Platau support that contingent local conditions drive recent treeline 559 patterns. Plant Ecol Divers 5:311–321. 560 Wang Y, Liang E, Sigdel SR, Liu B, Camarero JJ (2017) The couplng of treeline elevaton and 561 temperature is mediated by non-thermal factors on the Tibetan Plateau. Forests 8:109. 21 562 Wang Y, Pederson N, Ellison AM, Buckley HL, Case BS, Liang E, Camarero JJ (2016a) Increased 563 stem density and competition may diminish the positive effects of warming at alpine treeline. 564 Ecology 97:1668–1679. 565 Wang Y, Zhu H, Liang E, Camarero JJ (2016b) Impact of plot shape and size on the evaluation of 566 treeline dynamics in the Tibetan Plateau.Trees 30:1045–1056. 567 Yang B, Bräuning A, Liu J, Davis ME, Shao Y (2009) Temperature changes on the Tibetan Plateau 568 during the past 600 years inferred from ice cores and tree rings. Glob Planet Chang 69:71–78. 569 Yao Q, Brown P M, Liu S, Rocca ME, Trouet V, Zheng B, Chen H, Li Y, Liu D, Wang X (2017) 570 Pacific-Atlantic Ocean influence on wildfires in northeast China (1774 to 2010). Geophys Res 571 Lett 44: 1025–1033. 572 Zhu H, Shao X, Xu P, Xu Y, Tian H (2011) August temperature variability in the 573 southeastern Tibetan Plateau since AD 1385 inferred from tree rings. 574 Palaeogeogr Palaeoecl Paleoecol 305: 84–92. 575 Zhu H, Zheng Y, Shao X, Liu X, Xu Y, Liang E (2008) Millennial temperature reconstruction based 576 on tree-ring widths of Qilian juniper from Wulan, Qinghai Province, China. Chin Sci Bull 53: 577 3914–3920. 578 Ziegenhagen LL, Miller RF (2009) Postfire recovery of two shrubs in the interiors of large burns 579 in the Intermountain West, USA. West. N Am Naturalist 69: 195–205. 22 580 581 Table 1. Description of the five study plots used to characterize treeline or timberline dynamics in 582 the Baima Snow Mountains, southeastern Tibetan Plateau. Plot code, plot size, latitude (Lat), 583 Longitude (Lon), elevation of current treeline (ECT), slope, fire disturbance (FD), tree desnity 584 (No, number of trees per hectare) are shown in the table. Plot Plot Tree Lat Lon ECT (m) Slope FD No code size species (N) (E) (º) (m) E1 30 × 500 Larix 28°21.308' 99°01.349' 4359 5 Yes 163 potaninii NE1 30 × 280 Larix 28°21.374' 99°01.494'’ 4417 19 Yes 544 potaninii NE2 30 × 270 Larix 28°21.557' 99°01.239' 4387 19 Yes 323 potaninii NE3 30 × 200 Abies 28°20.955' 99°01.912' 4247 25 No 1295 georgei N1 30 × 200 Larix 28°20.899' 99°01.767' 4305 21 No 188 potaninii 585 586 23 587 Table 2. Significance levels of two-sample Kolmogorov-Smirnov (K-S) tests showing the 588 difference between recruitment series within the five study plots. Recruitment data of plot pairs Significance Disturbance regimes level of the K- S test (p) E1 vs. NE1 0.576 Fire-disturbed vs. fire-disturbed E1 vs. NE2 0.099 Fire-disturbed vs. fire-disturbed NE1 vs. NE2 0.517 Fire-disturbed vs. fire-disturbed E1 vs. N1 0.0041 Fire-disturbed vs. undisturbed E1 vs. NE3 0.0011 Fire-disturbed vs. undisturbed NE1 vs. N1 0.0032 Fire-disturbed vs. undisturbed NE1 vs. NE3 0.0003 Fire-disturbed vs. undisturbed NE2 vs. N1 0.0019 Fire-disturbed vs. undisturbed NE2 vs. NE3 0.0041 Fire-disturbed vs. undisturbed NE3 vs. NE 0.586 Undisturbed vs. undisturbed 589 24 590 591 Table 3. Comparison between treeline shifts during the three periods (1980-2014 vs. 1961-2014 and 1980-2014 vs. 1911-2014). The percentages were 592 calculated by dividing the treeline shifts inferred for 1980-2014 and 1961-2014 or 1911-2014, respectively. The vegetation thickness index (TI, plant height 593 × cover) was used to indicate the species interactions beyond the treeline. TIs < 0.8 and > 0.8 represent facilitation and competition, respectively Treeline Tree species and fire Mean shrub height Mean shrub cover TI Magnitude of treeline shift Percentage plot disturbance above treeline (m) above treeline (%) (m a.sl.) (%) 1911–2014 1961–2014 1980-2014 E1 Larch, disturbed (1) 0.45±0.07 90±3.8 0.41 unknown 12 11 91.7, unknown NE1 Larch, disturbed (1) 0.30±0.06 75±7.2 0.23 unknown 44 44 100, unknown NE2 Larch, disturbed (1) 0.75±0.07 89±5.1 0.67 22 22 22 100, 100 NE3 fir, undisturbed (0) 2.2±0.29 85±3.8 1.87 2 2 NA NA, NA N1 Larch, undisturbed (0) 2.0±0.26 82±6.5 1.64 4 NA NA NA, NA 594 (see more details in Liang et al., 2016). Periods without trees are shown as “unknown”. Fire occurrence was indicated by binary variable (1, 0). 595 596 25 597 598 Figures Caption 599 Figure 1. Location of the Baima Snow Mountains on the southeastern Tibetan Plateau (a), larch 600 (Larix potaninii) and fir (Abies georgei) treeline plots in the Baima Snow Mountains (photograph 601 by Y. Wang, October 2014) (b). The yellow and green areas in the image correspond to larch and 602 fir treelines, respectively. 603 604 Figure 2. Growth patterns for living and fire-killed larch trees forming the treeline. The grey and 605 black dashed lines denote the number of living and fire-killed trees, respectively. 606 607 Figure 3. Long-term changes in summer temperature and abrupt changes in tree recruitment at the 608 treeline. Temperature series was reconstructed from tree rings on the southeastern Tibetan Plateau 609 (SE TP) (Zhu et al., 2011) (a). Analyses of data for abrupt changes (black horizontal lines) in tree 610 recruitment reconstructed for the treelines (b-f). Shifts were detected in the 1990s (b-d), 1970s (e), 611 1900s and 1980s (f), corresponding to abrupt increases in tree recruitment. All study plots 612 correspond to larch treelines, except the NE3 plot which was located at a fir treeline. 613 614 Figure 4. Spatio-temporal variability in tree density and treeline position (uppermost elevation of 615 living trees ≥ 2 m) at five treeline plots located in the Baima Snow Mountains, southeastern 616 Tibetan Plateau. Closed symbos denote an individual that was established during the period 617 indicated at the top of each plot, whereas open symbols respresent trees established during 618 previous periods. Different symbols correspond to different tree establishment periods. 619 26 620 621 622 Figure 1 623 27 624 Fire-killed trees 3.2 Living trees 100 90 2.8 80 2.4 Mean tree-ring width (mm) 70 Number of trees 2.0 60 1.6 50 40 1.2 30 0.8 20 0.4 10 0.0 0 1700 1750 1800 1850 1900 1950 2000 625 Year 626 Figure 2 627 28 628 629 Abrupt change detection (A) Summer mean temperature on the SE TP 2 Z-score 0 -2 120 (B) Plot E1: larch, fire-disturbed 80 40 0 300 (C) Plot NE1: larch, fire-disturbed 200 100 Number of trees 0 150 (D) Plot NE2, larch, fire-disturbed 100 50 0 200 (E) Plot NE3: fir, undisturbed 150 100 50 0 16 (F) Plot N1: larch, undisturbed 12 8 4 0 1711-1720 1751-1760 1771-1780 1831-1840 1851-1860 1871-1880 1951-1960 1971-1980 1731-1740 1811-1820 1891-1900 1911-1920 1931-1940 1991-2000 1791-1800 Decade 630 631 Figure 3 632 29 (A) Plot N1: larch, undisturbed 1711-1760 1761-1810 1811-1860 1861-1910 1911-1960 1961-2013 200 4299 m 4301 m 4305m 175 Tree line 150 4282 m 125 Y (m) Increasing altitude (m) 100 75 50 25 0 0 30 30 30 30 30 30 X (m) 633 (B) NE3: fir, undisturbed 1711-1760 1761-1810 1811-1860 1861-1910 1911-1960 1961-2012 200 Treeline 4247 m 175 4245 m 150 Y (m) Increasing altitude (m) 125 100 75 50 25 0 0 30 30 30 30 30 30 X (m) 634 635 (Continued) 636 30 (C) E1: larch, fire-disturbed 1811-1860 1861-1910 1911-1960 1961-2014 500 1711-1760 1761-1810 4359 m 400 Treeline 4347 m Y (m) Increasing altitude (m) 300 200 100 0 0 30 30 30 30 30 30 X (m) 637 (D) NE1: larch, fire-disturbed 1711-1760 1761-1810 1811-1860 1861-1910 1911-1960 1961-2014 280 4417 m 240 Y (m) Increasing altitude (m) 200 160 4373 m 120 80 40 0 0 30 30 30 30 30 30 X (m) 638 639 (Continued) 640 31 (E) NE2: larch, fire-disturbed 1811-1860 1911-1960 1961-2014 270 1711-1760 1761-1810 1861-1910 240 210 Y (m) Increasing altitude (m) 180 4387 m 150 120 90 4365 m 4343 m 60 Treeline 4334 m 30 0 0 30 30 30 30 30 30 641 X (m) 642 Figure 4 643 32 644 645 Supplementary information 646 Table S1. Information about eight larch (Larix potaninii) seedlings located above current larch 647 treelines in the Baima Snow Mountains, southeastern Tibetan Plateau. Seedling Latitude (N), Altitude Height Age Nearby plot Longitude (E) (m a.s.l) (m) (years) 1 28°21.737', 99°01.215', 4416 0.60 25 E1 2 28°21.181', 99°01.182' 4416 0.35 18 E1 3 28°21.232', 99°01.178' 4435 0.40 19 E1 4 28°21.286', 99°0.955' 4499 0.40 26 E1 5 28°21.364', 99°01.158' 4448 0.67 23 NE1 6 28°21.642', 99°01.368' 4482 0.65 24 NE1 7 28°21.301', 99°0.945' 4501 0.65 21 NE1 8 28°21.313', 99°0.956' 4505 0.24 32 NE2 648 649 33 650 651 Figure S1 Ages of seedlings obtained from counting the internodes and the rings of stem disks at 652 the treeline in the Baima Snow Mountains. Counting rings of stem disks 24 Counting internodes 20 Age of seedlings (yr) 16 12 8 0 5 10 15 20 25 30 Mark number for samples 653 654 34