This is the accepted version of the journal article: Fang, Chao; Verbrigghe, Niel; Sigurdsson, Bjarni D.; [et al.]. «Decadal soil warming decreased vascular plant above and below ground production in a subarctic grassland by inducing nitrogen limitation». New Phytologist, Vol. 240, issue 2 (Oct. 2023), p. 565-576. DOI 10.1111/nph.19177 This version is available at https://ddd.uab.cat/record/284316 under the terms of the license 1 Decadal soil warming decreased vascular plant above- and below-ground 2 production in a subarctic grassland by inducing nitrogen limitation 3 Chao Fanga, b*, Niel Verbriggheb, Bjarni D. Sigurdssonc, Ivika Ostonend, Niki 4 Leblanse, Sara Marañón Jiménezf, g, h, Lucia Fuchsluegeri, Páll Sigurðssonc, Kathiravan 5 M. Meeranj, Miguel Portillo Estradab, Erik Verbruggenb, Andreas Richteri, Jordi 6 Sardansf, g, Josep Peñuelasf, g, Michel Bahnj, Sara Viccab, Ivan A. Janssensb 7 a 8 Research Center for Global Changes and Ecosystem Carbon Sequestration & 9 Mitigation, School of Applied Meteorology, Nanjing University of Information Science 10 and Technology, Nanjing 210044, China b 11 PLECO (Plants and Ecosystems), Department of Biology, University of Antwerp, 12 Universiteitsplein 1, 2610 Wilrijk, Belgium c 13 Agricultural University of Iceland, Hvanneyri, IS-311, Borgarnes, Iceland. d 14 Institute of Ecology and Earth Sciences, University of Tartu, Tartu 51003, Estonia e 15 Climate Impacts Research Centre, Umeå University, Umeå 90333, Sweden. f 16 CREAF, Cerdanyola del Vallès, Barcelona 08193, Catalonia, Spain g 17 CSIC, Global Ecology Unit CREAF‐ CSIC‐ UAB, Bellaterra, Barcelona 08193, 18 Catalonia, Spain h 19 Universitat Autònoma de Barcelona, Cerdanyola del Vallès 08193, Spain i 20 Centre for Microbiology and Environmental Systems Science, University of Vienna, 21 Djerassiplatz 1, 1030 Vienna, Austria j 22 Department of Ecology, University of Innsbruck, 6020 Innsbruck, Austria 23 24 *Corresponding author: Full telephone: +86 18293191870, Fax No.: 86-25-57792648, 25 E-mail address: fangchao567@gmail.com 26 Abstract 27 The below- and aboveground dynamics of vegetation are crucial in understanding how 28 climate warming may affect terrestrial ecosystem carbon cycling. In contrast to above- 29 ground biomass, the response of below-ground biomass to long-term warming has been 30 poorly studied. Here, we characterized the impacts of decadal geothermal soil warming 31 at two levels (on average +3.3 ºC and +7.9 ºC) above ambient soil temperature on 32 above- and below-ground plant biomass stocks and production in a subarctic grassland. 33 We tested the following two hypotheses: 1) shoot and fine root production both increase 34 with long-term soil warming in response to the more benign climate and the accelerated 35 rate of nutrient cycling; 2) because soil nitrogen (N) availability typically increases in 36 warmed mesic and cold ecosystems, the root-shoot ratio decreases with long-term soil 37 warming, resulting in warming-induced above-, but not below-ground increases in plant 38 biomass. Both hypotheses were rejected: long-term soil warming did not change 39 standing root biomass and even decreased fine root production, and also negatively 40 affected aboveground biomass and production. Decadal soil warming also did not 41 statistically significantly alter the root-shoot ratio. Structural equation modelling 42 suggested that following 10 years of soil warming, temperature was no longer the direct 43 driver of the observed changes, but a change in soil N was. Soil warming-induced 44 decreases in soil organic matter and water retention capacity indicated stronger 45 eluviation, which could explain soil N loss. The decreased shoot production was 46 attributed to decreased soil N concentration. The reduction in fine root production was 47 also related to the decreased soil N stocks, albeit indirectly via a decrease in specific 48 root area. Furthermore, soil N limitation induced by the decadal soil warming increased 49 the moss-vascular plant ratio in above-ground biomass, implying a setback in the plant 50 community succession towards a community more dependent on fresh nutrient inputs 51 from atmospheric deposition in the subarctic grassland in a warmer climate change. 52 These results indicate that after a decade of soil warming, plant productivity in the 53 studied subarctic grassland ecosystems were affected by soil warming mainly by the 54 reduction in soil N. 55 Keywords: vascular plants; grasses; biomass distribution; temperature increase; N 56 limitation. 57 Introduction 58 Climate warming induced by increasing atmospheric greenhouse gases has increased 59 the global mean temperature by approximately 1 °C since the industrial revolution, and 60 this warming is even more pronounced at high latitudes (IPCC, 2018). Climate warming 61 promotes soil organic matter (SOM) decomposition, releasing CO2 to the atmosphere 62 that may elicit a positive ecosystem feedback when not compensated for by increased 63 vegetation growth and carbon (C) uptake and storage in biomass or soils (Crowther et 64 al., 2019; Davidson & Janssens, 2006; Luo, Wan, Hui, & Wallace, 2001; Melillo et al., 65 2002; Raich & Schlesinger, 1992), especially at high latitudes (Dorrepaal et al., 2009; 66 Walker et al., 2020). Above-ground litter fall and root turnover constitute the primary 67 input of organic matter into soil, in addition to root exudates and mycorrhizae (Godbold 68 et al., 2006; Ven, Verlinden, Verbruggen, & Vicca, 2019; Vicca et al., 2012). Therefore, 69 understanding the responses of above- and below-ground production to projected 70 climate warming is relevant for both a better understanding of ecosystem function in a 71 changing environment, and more accurate quantification of the climate-carbon 72 feedback by Earth system models. Knowledge on the effects of warming on below- 73 ground plant dynamics, however, is very limited compared to the much better 74 understood above-ground production, despite its substantial contribution to ecosystem 75 productivity, especially in northern ecosystems (Ottaviani et al., 2020; Kong et al., 2019; 76 Qi, Wei, Chen, & Chen, 2019; See et al., 2019). 77 In addition to being a key component of global plant production (Ma et al. 2021) and 78 ecosystem C cycle (Bardgett et al. 2014), root- and rhizosphere inputs into the soil 79 contribute more to soil organic matter (SOM) formation than above-ground inputs 80 (Godbold et al., 2006; Kätterer et al., 2011; Slessarev et al., 2020), although root inputs 81 were also shown to elicit carbon losses by priming the rhizosphere (Guenet et al., 2018; 82 Dijkstra et al., 2021). In particular, in high-latitude ecosystems, where the supply of 83 nutrients, especially nitrogen (N), is low (Salazar et al., 2020), below-ground 84 investment is more pronounced (Iversen et al., 2015) and often more important in 85 ecosystem C balance than the above-ground litter inputs (Crowther et al., 2019; 86 Semchenko et al., 2018; Walker et al., 2018). Fine roots (defined here as roots with a 87 diameter < 2 mm) are the primary pathway for the uptake of water and mineral nutrients 88 to support plant growth, thereby linking below- and above-ground C processes (Gao et 89 al., 2008; McCormack et al., 2015; See et al., 2019). Fine root growth accounts for 90 roughly 20% of global terrestrial net primary production (McCormack et al., 2015) and 91 is very sensitive to environmental changes, particularly to warming (Bai et al., 2010). 92 Published responses of grassland fine root production to warming have been 93 inconsistent, with positive (Arndal, Tolver, Larsen, Beier, & Schmidt, 2017), negative 94 (Bai et al., 2010) or neutral (Schweiger et al., 2018) responses all having been observed. 95 In drylands, warming can be detrimental by increasing aridity, but in cold and mesic 96 ecosystems, such as studied here, warming tends to ameliorate conditions for plant 97 growth and may thus lead to increased shoot and root production (Wang et al. 2019). 98 Warming also accelerates SOM decay and nutrient mineralization, thereby reducing the 99 need to invest in roots to sustain high above-ground productivity (Dieleman et al., 2012). 100 The lack of consistency in the warming effect on fine roots in the literature may thus be 101 attributed to the fact that fine root production can be influenced by warming in various 102 direct and indirect ways (Song et al., 2019). In addition to altered soil water and N 103 availability, microbial enzymatic activities and above-ground productivity have also 104 been suggested as determinants of fine root production (Dybzinski et al., 2019; Fortier, 105 Truax, Gagnon & Lambert, 2019; Jourdan et al., 2008; Ma et al., 2012; Peek et al., 106 2006; Rygiewicz & Andersen, 1994). Therefore, it is relevant to explore the 107 mechanisms underlying fine root production in response to future global warming. 108 Warming may further alter the distribution of biomass in plants. Biomass distribution 109 is a very important concept in understanding below- and above-ground functions and 110 in modelling the changes in community structure and ecosystem function under climate 111 change (Poorter et al. 2012). The most commonly used approach to study changes in 112 biomass distribution is to measure the root-shoot ratio (defined as root biomass divided 113 by above-ground biomass), which reflects the relative difference in the response 114 strategies of below- and above-ground tissues to warming and resource availability 115 (Mokany et al. 2006; Song et al. 2019). The functional equilibrium model suggests that 116 plants can shift the allocation of C to the below- or above-ground parts based on 117 resource constraints (Bloom et al., 1985). For example, plants allocate more C to above- 118 ground organs in resource-rich soil and more C to their roots in resource-poor 119 environments (Bloom, Chapin, & Mooney, 1985; Kobe et al., 2010). Nitrogen is 120 considered a key element in regulating plant productivity in Northern areas (Du et al., 121 2020; Kou et al., 2020; LeBauer & Treseder, 2008; Myrstener et al., 2018; Penuelas et 122 al., 2013; Thomas, et al, 2013). In the absence of drought, warming generally 123 accelerates the decomposition of SOM, thus releasing more N into the soil and 124 improving soil N availability (Salazar et al., 2020). Therefore, increased N availability 125 induced by warming is expected to promote plant growth and shift C allocation to 126 above-ground parts, subsequently leading to a reduction in the root-shoot ratio, 127 especially in cold grasslands where N availability is low. 128 To study how warming affects below- and above-ground biomass and productivity, 129 temperature manipulation experiments have been widely conducted (Nijs et al., 1996; 130 Xu et al., 2012; Fang et al., 2017; Fang et al., 2018; Maestre et al., 2013; Melillo et al., 131 2011; Pries et al.,2017). However, a critical drawback of most warming experiments is 132 their short-term nature. Given that the impact of climate warming on ecosystems 133 includes ecosystem state responses that may take decades to equilibrate (Melillo et al., 134 2017; Walker et al., 2020), such as changes in soil structure or vegetation community 135 composition, short-term experiments may not be representative of how ecosystems will 136 respond to future warming in the long term. 137 To evaluate the long-term effects of soil warming on below- and above-ground biomass 138 distribution and production, we conducted a study across soil temperature gradients 139 created by geothermal activity in a subarctic grassland in Iceland, located in a humid 140 oceanic region, that started 10 years earlier. Two hypotheses were tested: (1) decadal 141 warming stimulates both shoot and fine root production, and 2) the increased soil N 142 availability induced by decadal warming has led to a decrease in the root-shoot ratio of 143 this subarctic grassland. 144 145 2 Materials and methods 146 2.1 Study area 147 The experiment was conducted in unmanaged grasslands at the ForHot research site in 148 the Hengil geothermal area (Sigurdsson et al., 2016), 40 km east of Reykjavik, Iceland 149 (64o00’01”N, 21o11’09”W; 83-168 m altitude). The area has an oceanic climate, 150 characterized by a mean annual air temperature of 5.2 ºC, with July being the warmest 151 month (12.2 ºC) and December the coldest (-0.1 ºC) (Icelandic Meteorological Office; 152 www.vedur.is). The mean annual precipitation and wind speed were 1460 mm and 6.6 153 m s-1, respectively (Icelandic Meteorological Office; www.vedur.is). The soils were 154 classified as Brown Andosols and had a silty-loamy texture (Sigurdsson et al., 2016). 155 The grassland, which was unmanaged but fenced to protect from livestock grazing, was 156 dominated by Agrostis capillaris, Ranunculus acris and Equisetum pratense, all 157 perennial species with short above-ground tissues that regrow each year from 158 underground stem or rhizomes (Leblans et al., 2017). 159 The soil at the study site had been warmed since May 2008, when an earthquake shifted 160 geothermal systems to previously unwarmed areas (Sigurdsson et al., 2016). Soil 161 warming occurs through horizontal heat conduction from fissures in the bedrock that 162 are warmed by penetrating hot groundwater. Geothermal water remains confined within 163 the bedrock and no signs of soil contamination were found by geothermal by-products, 164 such as exchangeable sulfur. Since hot water intrudes the bedrock through faults, 165 horizontal temperature gradients occur, with higher temperatures near the fault, and 166 declining temperatures perpendicular to the fault. Further detailed description of the 167 study site can be found in (Sigurdsson et al., 2016). 168 2.2 Experimental design 169 Five temperature transects were established in autumn 2012, with soil warming at 10 170 cm depth ranging from +0 ºC to +15.9 ºC, where six 2×2 m permanent plots were 171 established along each transect at a different level of warming (Sigurdsson et al., 2016). 172 Soil temperature was recorded hourly at 10 cm soil depth in each permeant plot using 173 TidbiT v2 HOBO® data loggers (Onset Computer Corporation, Bourne, Massachusetts, 174 USA). Given the high workload associated with root studies, we selected three warming 175 levels in each of the five transects according to climate change scenarios. Soil 176 temperatures in the selected plots were on average +0 ºC, +3.3 ºC and +7.9 ºC above 177 ambient (mean annual temperatures from July 2017 to July 2018 at 10 cm depth; Table 178 1). 179 2.3 Soil properties 180 In July 2018, ten years after the onset of the soil warming, the soils were sampled to 181 measure total soil N and C, and soil bulk density using a cutting ring (4.6 cm inner- 182 diameter) (Kleibl, Klvac, Lombardini, Porhaly, & Spinelli, 2014). Total soil N and C 183 were determined by dry combustion with a Thermo Flash 2000 NC Analyzer (Thermo 184 Fisher Scientific, Delft, The Netherlands). In July 2014, two soil cores in each plot were 185 taken to a depth of 10 cm using an auger (4.6 cm inner-diameter), one to measure total 186 extractable phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg) and one 187 to measure soil bulk density using a cutting ring (4.6 cm inner-diameter) (Kleibl, Klvac, 188 Lombardini, Porhaly, & Spinelli, 2014). A 7.5 g subsample of fresh soil was extracted 189 with 0.5 M NaHCO3 within 24 h of sampling, digested at 400 ºC with H2SO4 and 190 selenium as catalyst, and total extractable P concentration was determined from the 191 digested NaHCO3 extracts (Marañón-Jiménez et al., 2019). Base cations (K+, Ca2+, 192 Mg2+) were measured using acid digestion with H2SO4, salicylic acid, H2O2 and the 193 selenium method (Courchesne, Turmel, & Beauchemin, 1996; Mautner, 1999). All 194 analyzes were performed by colorimetric detection with a San++ Continuous Flow 195 Analyzer (Skalar Analytical B.V., Breda, The Netherlands). Nutrient pools were 196 calculated as the concentration of soil nutrients multiplied by soil bulk density and soil 197 depth. 198 2.4 Plant properties 199 In July 2018, above- and below-ground plant biomass stocks were measured. Above- 200 ground biomass was measured by clipping a 0.2×0.5 m area to the soil surface in each 201 plot. In the same clipped area, two soil cores were taken to a depth of 15 cm using an 202 auger (4.6 cm inner-diameter) to measure standing root biomass. Above-ground 203 samples were sorted into vascular plants and mosses and dried to constant mass at 70 204 ºC. Oven dried above-ground vascular plant biomass is referred as both shoot biomass 205 and production hereafter, because all vascular species lose their above-ground parts 206 during winter and had reached their maximum biomass at the end of July when it was 207 measured. Standing biomass of moss was divided by above-ground standing biomass 208 of vascular plants to estimate the moss/vascular plant ratio as an indicator of plant 209 species composition change. 210 Standing root biomass was estimated from the soil cores. Roots were carefully washed 211 by wet sieving in a 0.15 mm mesh under gently flowing water to remove attached soil 212 and were subsequently dried to constant mass at 70 ºC and weighed to quantify root 213 standing biomass. Fine root production was measured by the root mesh method (Hirano 214 et al., 2009). In April 2018 (growing season starts in late May) (Leblans et al., 2017), 215 in each plot two root meshes with 2 mm mesh size, 10 cm length and 7.5 cm width were 216 vertically inserted down to 10 cm depth in the soil using a straight stainless steel blade 217 with 10 cm width, 20 cm length and 2 mm thickness. At the end of September in 2018 218 (the root growing season stops in August as observed in minirhizotrons), the root 219 meshes were extracted from the soil as 3×7.5 cm soil blocks up to 10 cm deep by 220 pushing two 7.5 cm wide and 20 cm long sharp straight and perching stainless steel 221 blades on 1.5 cm apart of both side of the mesh. The root meshes were missed in one 222 plot of +0 ºC treatment. The collected soil volume was then processed to obtain the wet 223 roots that passed through the mesh. The total dry mass of all roots that grew through 224 the mesh between April to September was considered as the annual fine root production, 225 assuming no further fine root productivity when aboveground tissues had senesced. 226 A subsample of about 0.07g (fresh mass) fine roots per plot was randomly taken from 227 the wet root samples, placed completely flat without overlap and scanned in the HP 228 ScanJet G2410 (HP Inc., United States). All of the scanned fine roots images were 229 analyzed using the software WinRHIZO Tron MF 2018b (Regent Instruments Inc., 230 Quebec, Canada) to quantify fine root area. Then the fine roots were dried to constant 231 mass at 70 ºC. The specific root area (SRA) was calculated by dividing the total fine 232 root area by the dried biomass of the root subsamples (Lõhmus, Oja, & Lasn, 1989). 233 The remaining roots were then dried to constant mass to obtain total root biomass. Root 234 turnover was calculated as the ratio of fine root production to standing root biomass. 235 2.6 Statistical analysis 236 Prior to statistical analysis, we assessed the data quality with Dixon’s Q test to identify 237 and remove outliers (Dixon, 1950; Efstathiou, 2006; Onoz & Oguz, 2003) to produce 238 the final dataset (Table A1 and A2). Specifically, three data in shoot production, one 239 data in total soil N, one data in soil P, one data in SRA were removed. Then, a mixed 240 linear model was applied to test the effects of soil warming on standing root biomass, 241 fine root and above-ground productivity, root-shoot ratio, soil water, soil bulk density, 242 and soil nutrients, with the three soil warming levels (i.e., +0 ºC, +3.3 ºC and +7.9 ºC) 243 as a fixed factor and transect as a random factor. Tukey-Kramer post-hoc tests in a 244 general linear model were applied due to the unequal sample sizes among treatments 245 after removing outliers, to determine the differences among treatments. Data were log- 246 transformed when required to ensure normality and homoscedasticity (Quinn & 247 Keough, 2009). Structural equation modelling (SEM) was performed using AMOS 21.0 248 to quantify the relative importance of the potential direct and indirect pathways in 249 mediating the soil warming effects (the real measured mean soil temperature) on fine 250 root and shoot production. First, a conceptual model was conceived based on plausible 251 relationships among variables (Fig. A1). Then the conceptual model was optimized by 252 Pearson’s correlation analysis based on the significant relationships. Specifically, the 253 paths without significant relationship were removed. Finally, the final structural 254 equation model was determined based on the goodness of model fit and logical 255 reasoning. All statistical analyses were performed using SPSS 21.0 (SPSS Inc., Chicago, 256 IL, USA). 257 258 3 Results 259 3.1 Effects of soil warming on below- and above-ground production and plant 260 traits 261 Fine root production was significantly lower in the warmed treatments than under 262 ambient temperature conditions, with the lowest fine root production in the +7.9 ºC 263 treatment (p < 0.05, Fig 1). However, there were no statistically significant differences 264 in standing root biomass, root turnover and fine root N concentration among the three 265 treatments (p > 0.05, Fig. 2B, C, and F), likely due to the high spatial heterogeneity 266 among plots. Total fine root area significantly declined with soil warming (Fig. A2), 267 which was remarkable given that SRA was significantly higher in the warmed 268 treatments than in the control treatment, with the highest value at +7.9 ºC (Fig. 2E) and 269 the decline in standing root biomass was not statistically significant (p > 0.05). 270 Alongside the reduced below-ground productivity, also the above-ground productivity 271 of vascular plants declined (p < 0.05, Fig 2A). Although there was a clear tendency for 272 a warming-induced increase in root-shoot ratio (Fig. 2D; vascular plant above-ground 273 biomass tended to decline less than fine root biomass), this increase was not statistically 274 significant (p > 0.05). There were no significant differences in the moss biomass among 275 +0 ºC, +3.3 ºC and +7.9 ºC treatments (Fig. A3). The moss-vascular plant ratio was 276 significantly higher in the warmed treatments than under ambient temperature (p < 0.05, 277 Fig 2A). 278 3.2 Effects of soil warming on soil water, bulk density and nutrients 279 Soil warming significantly decreased soil water content (Fig. 3A and B), but water 280 content remained high. Soil bulk density was significantly higher under +7.9 ºC 281 treatment than +3.3 ºC treatment and the control treatment, while no significant 282 difference was observed between the +3.3 ºC treatment and the control treatment (Fig. 283 3C). Both total soil N concentration and pool were significantly lower in the warmed 284 treatments than in the control treatment (p < 0.05, Fig. 4A and F). Soil warming 285 significantly decreased the soil P concentration (p < 0.05, Fig. 4B), but did not change 286 soil P pool (Fig. 4G). No significant warming effects were observed in total soil 287 extractable K+, Ca2+ and Mg2+ concentrations or pools (Fig. 4C-E and H-J). 288 3.3 Relationships of below- and above-ground biomass and production with 289 potential drivers 290 Shoot biomass (and thus production) and fine root production were significantly 291 negatively correlated with measured mean annual soil temperature, moss-vascular plant 292 ratio and SRA (p < 0.05, Table 2, Fig. 5A, C and E) and significantly positively 293 correlated with soil N concentration (p < 0.05, Table 2, Fig. 5D), but not with soil bulk 294 density (p > 0.05, Table 2, Fig. 5b and G). Fine root production correlated negatively 295 with soil bulk density (p < 0.05, Table 2, Fig. 5L), but no significant relationship 296 between shoot production and soil bulk density was observed (p > 0.05, Table 2, Fig. 297 5B). Standing root biomass exhibited a significant negative relationship with soil 298 temperature (p < 0.05, Table 2, Fig. 5F), but not with soil bulk density, Moss/Vascular 299 plant, soil N concentration or SRA (p > 0.05, Table 2, Fig. 5G-J). 300 3.4 Linear stepwise regression modelling and structural equation modelling of fine 301 root production and shoot production 302 The SEM analysis for fine root production suggested that soil warming affected fine 303 root production, but only indirectly through increasing soil bulk density and reducing 304 soil N concentration, which together triggered the increase in SRA that drove the lower 305 fine root production (Fig. 6). Amongst all direct and indirect effects, the SRA was the 306 most important predictor directly shaping the variation in fine root production (Table 3, 307 Fig. 6). The direct effects of the reduced competitiveness of the vascular plants under 308 these conditions (increased moss-vascular plant ratio), soil warming, soil nitrogen, and 309 bulk density, were all statistically insignificant (at p > 0.05), but combined the SEM did 310 explain 83% of the variance in fine root production (Fig. 6). The SEM thus revealed 311 that the abiotic changes in the warmed soil elicited morphological adaptations in the 312 fine root system that resulted in reduced fine root production. 313 The SEM of shoot production showed that soil warming indirectly affected shoot 314 production through increasing soil bulk density and reducing soil N concentration. 315 Together, soil N concentration was the most important predictor shaping the variation 316 in shoot production (Table 3 and Fig. 7). These abiotic variables explained 96% of the 317 variation in shoot production (Fig. 7). 318 319 4 Discussion 320 This study tested two hypotheses: 1) shoot and fine root production increase with long- 321 term soil warming, because of ameliorated growth conditions and accelerated N cycling 322 (Noyce, Kirwan, Rich, & Megonigal, 2019; Schaeffer, Sharp, Schimel, & Welker, 2013), 323 and 2) the root-shoot ratio decreases with long-term soil warming (optimal partitioning 324 theory). In contrast to these hypotheses, we found that the increase in soil temperature 325 reduced both below- and above-ground production similarly and did not change root- 326 shoot ratio (Figs. 1 and 2). Our results revealed that, in contrast to what was expected 327 (warming accelerating nutrient cycling and thereby stimulating productivity), warming 328 induced a loss of soil nitrogen and an increase in soil bulk density (both driven by a 329 substantial loss of SOM; Maranon-Jimenez, 2019; Verbrigghe et al., 2022), which 330 elicited a reduction in both below- and above-ground productivity. 331 Previous studies have shown that increased temperature typically increases above- 332 ground production (Dieleman et al., 2012; Fang et al., 2018; Sherry et al., 2008; Wan, 333 Hui, Wallace, & Luo, 2005; Xia & Wan, 2013), mainly by extending the growing season 334 (Fang et al., 2018; Wan et al., 2005; Xia & Wan, 2013), promoting photosynthesis 335 (Lewis, Lucash, Olszyk, & Tingey, 2011), and/or increasing soil N availability (Sherry 336 et al., 2008). A previous study at the same site suggested that annual photosynthesis 337 (carbon input) might have increased in warmer plots because the integrated seasonal 338 NDVI increased at higher temperatures (Leblans et al., 2017). However, Callebaut 339 (2022) showed that leaf-level photosynthetic capacity in one of the dominant vascular 340 plant species was reduced in warmed plots, due primarily to reduced foliar N 341 concentrations. Nitrogen has indeed been identified to limit grassland productivity in 342 Iceland (Leblans et al., 2014; Leblans et al., 2017), and in many other northern areas 343 with low atmospheric N deposition (Liu et al., 2020). The strong decline in soil N-P 344 ratio indeed suggests that N limitation in these grasslands aggravated in response to 345 decadal warming (Fig. A4). 346 Previous studies at the site showed that soil warming accelerated SOM decomposition 347 (Marañón-Jiménez et al., 2018; Walker et al., 2018), resulting in decreased SOM stocks 348 (Verbrigghe et al., 2022), while the soil C-N ratio remained unaltered, suggesting that 349 plants and microbes were not able to retain the mineralized N within the ecosystem. 350 Warming thus induced substantial losses of soil N (Fig. 3A and F, Marañón-Jiménez et 351 al., 2019) that explained the reduced productivity, both aboveground as below-ground. 352 Soil organic matter is positively correlated with soil water retention (Rawls et al., 2003). 353 The warming-induced decrease of SOM thus likely resulted in lower soil water 354 retention capacity (Fig.3A and B), and thereby in increased water drainage and N 355 leaching out of the ecosystem. Increased leaching and reduced N retention capacity by 356 the smaller microbial biomass likely explain the strongly depleted soil N pool 357 (Marañón-Jimenez et al. 2018; Walker et al. 2020), driving reduced N availability for 358 plant growth at the site. The SEM indeed suggested that N limitation induced by 359 warming directly decreased shoot production. Also the linear stepwise regression model 360 indicated that soil N concentration was the main driver of shoot production. Therefore, 361 we conclude that the decreased soil N concentration induced by decadal warming and 362 accelerated N losses, explains the decrease of above-ground plant productivity. 363 In parallel with aboveground productivity, also fine root productivity was expected to 364 increase with warming in cold grasslands, where roots could benefit from longer 365 growing seasons (Price and Waser 1998) and higher N mineralization rates (Wang et al. 366 2019) induced by warming. By contrast, decadal soil warming at our study site 367 decreased fine root production. In contrast to aboveground productivity, however, the 368 SEM showed no statistically significant direct contribution of the reduced N availability 369 to the decreased fine root production. In addition, while soil warming was shown to 370 advance the onset of the growing season (Fig. A, Leblans et al., 2017), the SEM also 371 did not suggest a direct effect of soil warming on fine root production. In contrast, both 372 the linear stepwise regression model and the SEM suggested that the decrease in fine 373 root production under warming was mainly attributable to the increased SRA, which 374 was induced by the decreased soil N concentration and the increased soil bulk density 375 (Table 2 and Fig. 7). The higher SRA indicates thinner roots and thus a larger root-soil 376 contact area, favoring N uptake (Hong, Ma, Yan, Zhang, & Wang, 2017). Thinner roots 377 are also better adapted to soils with increased bulk density and smaller pores, as 378 observed in our study site, and come at reduced construction costs. Nonetheless, the 379 increased SRA and associated lower C cost and enhanced N uptake efficiency did not 380 suffice to sustain aboveground productivity. 381 In contrast to our second hypothesis, no significant effect of soil warming on root-shoot 382 ratio was observed in this study. Temperature is an important limiting factor for plant 383 growth in cold ecosystems (Sistla et al., 2013). Increased soil temperatures could thus 384 have induced a more favorable soil thermal environment for vegetation through earlier 385 onset of the growing season in this subarctic grassland (Fig. A5, Leblans et al., 2017). 386 Bai et al. (2010) suggested that the observed responses of plant C allocation to warming 387 resulted from the balance between favorable and unfavorable environments in a 388 temperate steppe. In our study, the reduced soil N stock likely counteracted this positive 389 effect of the more benign soil thermal environment on vascular plant production, 390 leading to no changes in root-shoot ratio in response to soil warming. These results 391 imply that the observed root-shoot ratio following long-term soil warming is the result 392 of a fine balance between potentially large and competing effects. 393 In conclusion, our in-situ soil warming study provided unique data to elucidate the 394 responses of below- and above-ground plant biomass productivity to decadal soil 395 warming. Our results showed that decadal soil warming did not change standing root 396 biomass or root-shoot ratio, but decreased aboveground and fine root production. 397 Decreased fine root production induced by soil warming was directly related to 398 increased SRA, for which increased soil N limitation and soil bulk density were the 399 most important regulating factors. Soil N limitation was also identified as the main 400 driver of the decreased vascular plant shoot production. Furthermore, decadal soil 401 warming increased moss-vascular plant ratio in above-ground biomass, implying a 402 setback in the plant community succession towards a community more dependent on 403 nutrients from atmospheric deposition and mineral weathering in the sub-arctic 404 grassland ecosystem in a warmer climate. These results indicate that after a decade of 405 soil warming, below- and aboveground plant productivity in this sub-arctic grassland 406 ecosystem was more affected by a warming-induced change in soil N than by the 407 warming per se. 408 Acknowledgements 409 This research was supported by the Icelandic Research Council (IRF Fund, No. 163272- 410 051 to BDS), and the MSCA ITN-Grant No 813114 FutureArtic, Estonian Research 411 Council grant PRG916. We thank the staff at the Agricultural University of Iceland, at 412 Reykir and Keldnaholt, who provided logistical support for the present study. We thank 413 Marc Wellens for making the sharp straight and perching stainless steel blade. We also 414 thank Inge Van De Putte, Jana Vynckier, Jochen Janssens, Maxime Sepelie and Iris 415 Janssens for helping in the collection and treatment of soil and plant samples. Further, 416 we thank Nadine Calluy for her assistance with the lab analyses. 417 Reference 418 Arndal, M. F., Tolver, A., Larsen, K. S., Beier, C., & Schmidt, I. K. (2017). Fine Root Growth and Vertical 419 Distribution in Response to Elevated CO2, Warming and Drought in a Mixed Heathland– 420 Grassland. Ecosystems, 21(1), 15-30. doi:10.1007/s10021-017-0131-2 421 Bai, W., Wan, S., Niu, S., Liu, W., Chen, Q., Wang, Q., . . . Li, L. (2010). Increased temperature and 422 precipitation interact to affect root production, mortality, and turnover in a temperate steppe: 423 implications for ecosystem C cycling. Global Change Biology, 16(4), 1306-1316. 424 doi:10.1111/j.1365-2486.2009.02019.x 425 Bardgett, R. D., L. Mommer, and F. T. De Vries. (2014). Going underground: root traits as drivers of 426 ecosystem processes. Trends Ecology and Evolution, 29, 692-699. 427 Bloom, A. J., Chapin, F. S., & Mooney, H. A. (1985). Resource limitation in plants: an economic analogy. 428 Annual Review of Ecology and Systematics, 16, 363–392. 429 doi:10.1146/annurev.es.16.110185.002051 430 Callebaut T. (2022). Gas exchange in subarctic grasslands. Impact of medium term and long term soil 431 warming on leaf gas exchange in subarctic Ranunculus acris L. MsC Thesis. University of 432 Antwerp. 433 Chen, J., Luo, Y., Xia, J., Wilcox, K. R., Cao, J., Zhou, X., . . . Wang, R.-W. (2016). Warming Effects on 434 Ecosystem Carbon Fluxes Are Modulated by Plant Functional Types. Ecosystems, 20(3), 515- 435 526. doi:10.1007/s10021-016-0035-6 436 Courchesne, F., Turmel, M., & Beauchemin, P. (1996). Magnesium and potassium release by weathering 437 in spodosols: Grain surface coating effects. Soil society of America Journal, 60(4), 1188-1196. 438 doi:10.2136/sssaj1996.03615995006000040033x 439 Crowther, T. W., van den Hoogen, J., Wan, J., Mayes, M. A., Keiser, A. D., Mo, L., . . . Maynard, D. S. 440 (2019). The global soil community and its influence on biogeochemistry. Science, 365(6455). 441 doi:10.1126/science.aav0550 442 Daebeler A, Bodelier PLE, Hefting MM, Rütting T, Jia Z, Laanbroek HJ. (2017). Soil warming and 443 fertilization altered rates of nitrogen transformation processes and selected for adapted 444 ammonia-oxidizing archaea in sub-arctic grassland soil. Soil Biology and Biochemistry 107: 445 114-124. 446 Davidson, E. A., & Janssens, I. A. (2006). Temperature sensitivity of soil carbon decomposition and 447 feedbacks to climate change. Nature, 440(7081), 165-173. doi:10.1038/nature04514. 448 De Boeck, H. J., Vicca, S., Roy, J., Nijs, I., Milcu, A., Kreyling, J., . . . Beier, C. (2015). Global Change 449 Experiments: Challenges and Opportunities. BioScience, 65(9), 922-931. 450 doi:10.1093/biosci/biv099 451 Dieleman, W. I., Vicca, S., Dijkstra, F. A., Hagedorn, F., Hovenden, M. J., Larsen, K. S., . . . Janssens, I. 452 A. (2012). Simple additive effects are rare: a quantitative review of plant biomass and soil 453 process responses to combined manipulations of CO 2 and temperature. Global Chang Biology, 454 18(9), 2681-2693. doi:10.1111/j.1365-2486.2012.02745.x 455 Dijkstra FA, Zhu B, Cheng W. (2021). Root effects on soil organic carbon: a double-edged sword. New 456 Phytologist, 230(1), 60-65. 457 Dixon, W. J. (1950). Analysis of extreme values. The Annals of Mathematical Statistics, 21(4), 488-506. 458 doi:https://www.jstor.org/stable/2236602 459 Dorrepaal, E., Toet, S., van Logtestijn, R. S. P., Swart, E., van de Weg, M. J., Callaghan, T. V., & Aerts, 460 R. (2009). Carbon respiration from subsurface peat accelerated by climate warming in the 461 subarctic. Nature, 460(7255), 616-619. doi:10.1038/nature08216 462 Du, E., Terrer, C., Pellegrini, A. F. A., Ahlström, A., van Lissa, C. J., Zhao, X., . . . Jackson, R. B. (2020). 463 Global patterns of terrestrial nitrogen and phosphorus limitation. Nature Geoscience, 13(3), 464 221-226. doi:10.1038/s41561-019-0530-4 465 Dybzinski, R., Kelvakis, A., McCabe, J., Panock, S., Anuchitlertchon, K., Vasarhelyi, L., . . . Farrior, C. 466 E. (2019). How are nitrogen availability, fine-root mass, and nitrogen uptake related empirically? 467 Implications for models and theory. Global Change Biology, 25(3), 885-899. 468 doi:10.1111/gcb.14541 469 Efstathiou, C. E. (2006). Estimation of type I error probability from experimental Dixon's "Q" parameter 470 on testing for outliers within small size data sets. Talanta, 69(5), 1068-1071. 471 doi:10.1016/j.talanta.2005.12.031 472 Eisenhauer, N., Cesarz, S., Koller, R., Worm, K., & Reich, P. B. (2012). Global change belowground: 473 impacts of elevated CO2, nitrogen, and summer drought on soil food webs and biodiversity. 474 Global Change Biology, 18(2), 435-447. doi:10.1111/j.1365-2486.2011.02555.x 475 Fang, C., Li, F., Pei, J., Ren, J., Gong, Y., Yuan, Z., . . . Ye, J.-S. (2018). Impacts of warming and nitrogen 476 addition on soil autotrophic and heterotrophic respiration in a semi-arid environment. 477 Agricultural and Forest Meteorology, 248, 449-457. doi:10.1016/j.agrformet.2017.10.032 478 Fang, C., Ye, J.-s., Gong, Y., Pei, J., Yuan, Z., Xie, C., . . . Yu, Y. (2017). Seasonal responses of soil 479 respiration to warming and nitrogen addition in a semi-arid alfalfa-pasture of the Loess Plateau, 480 China. Science the Total Environment, 590-591, 729-738. doi:10.1016/j.scitotenv.2017.03.034 481 Fortier, J., Truax, B., Gagnon, D., & Lambert, F. (2019). Abiotic and biotic factors controlling fine root 482 biomass, carbon and nutrients in closed-canopy hybrid poplar stands on post-agricultural land. 483 Scitific Reports, 9(1), 6296. doi:10.1038/s41598-019-42709-6 484 Gao, Y. Z., Giese, M., Lin, S., Sattelmacher, B., Zhao, Y., & Brueck, H. (2008). Belowground net primary 485 productivity and biomass allocation of a grassland in Inner Mongolia is affected by grazing 486 intensity. Plant and Soil, 307(1-2), 41-50. doi:10.1007/s11104-008-9579-3 487 Gargallo-Garriga, A., Ayala-Roque, M., Sardans, J., Bartrons, M., Granda, V., Sigurdsson, B. D., . . . 488 Penuelas, J. (2017). Impact of Soil Warming on the Plant Metabolome of Icelandic Grasslands. 489 Metabolites, 7(3). doi:10.3390/metabo7030044 490 Godbold, D. L., Hoosbeek, M. R., Lukac, M., Cotrufo, M. F., Janssens, I. A., Ceulemans, R., . . . 491 Peressotti, A. (2006). Mycorrhizal Hyphal Turnover as a Dominant Process for Carbon Input 492 into Soil Organic Matter. Plant and Soil, 281(1-2), 15-24. doi:10.1007/s11104-005-3701-6 493 Guenet, B., Camino-Serrano, M., Ciais, P., Tifafi, M., Maignan, F., Soong, J. L., & Janssens, I. A. (2018). 494 Impact of priming on global soil carbon stocks. Global Change Biology, 24(5), 1873-1883. 495 doi:10.1111/gcb.14069 496 Hirano, Y., Noguchi, K., Ohashi, M., Hishi, T., Makita, N., Fujii, S., & Finér, L. (2009). A new method 497 for placing and lifting root meshes for estimating fine root production in forest ecosystems. 498 Plant Root, 3, 26-31. doi:10.3117/plantroot.3.26 499 Hofer, D., Suter, M., Buchmann, N., & Lüscher, A. (2017). Severe water deficit restricts biomass 500 production of Lolium perenne L. and Trifolium repens L. and causes foliar nitrogen but not 501 carbohydrate limitation. Plant and Soil, 421(1-2), 367-380. doi:10.1007/s11104-017-3439-y 502 Holmstrup, M., Ehlers, B. K., Slotsbo, S., Ilieva-Makulec, K., Sigurdsson, B. D., Leblans, N. I. W., . . . 503 Rasmann, S. (2018). Functional diversity of Collembola is reduced in soils subjected to short- 504 term, but not long-term, geothermal warming. Functional Ecology, 32(5), 1304-1316. 505 doi:10.1111/1365-2435.13058 506 Hong, J., Ma, X., Yan, Y., Zhang, X., & Wang, X. (2017). Which root traits determine nitrogen uptake 507 by alpine plant species on the Tibetan Plateau? Plant and Soil, 424(1-2), 63-72. 508 doi:10.1007/s11104-017-3434-3. 509 IPCC (2018). Global Warming of 1.5 °C. An IPCC Special Report on the impacts of global warming of 510 1.5 °C above pre-industrial levels and related global greenhouse gas emission pathways, in the 511 context of strengthening the global response to the threat of climate change, sustainable 512 development, and efforts to eradicate poverty (eds Masson-Delmotte, V. et al) 513 https://www.ipcc.ch/sr15/. 514 Iversen CM, Sloan VL, Sullivan PF, Euskirchen ES, McGuire AD, Norby RJ, Walker AP, Warren JM, 515 Wullschleger SD. (2015). The unseen iceberg: plant roots in arctic tundra. New Phytologist, 516 205(1): 34-58. 517 Johnson, M. G., Rygiewicz, P. T., Tingey, D. T., & Phillips, D. L. (2006). Elevated CO 2 and elevated 518 temperature have no effect on Douglas-fir fine-root dynamics in nitrogen-poor soil. New 519 Phytologist, 170(2), 345-356. doi:10.1111/j.1469-8137.2006.01658.x 520 Jourdan, C., Silva, E. V., Gonçalves, J. L. M., Ranger, J., Moreira, R. M., & Laclau, J. P. (2008). Fine 521 root production and turnover in Brazilian Eucalyptus plantations under contrasting nitrogen 522 fertilization regimes. Forest Ecology and Management, 256(3), 396-404. 523 doi:10.1016/j.foreco.2008.04.034 524 Kätterer T, Bolinder MA, Andrén O, Kirchmann H, Menichetti L. (2011). Roots contribute more to 525 refractory soil organic matter than above-ground crop residues, as revealed by a long-term field 526 experiment. Agriculture, Ecosystems & Environment 141(1), 184-192. 527 Kleibl, M., Klvac, R., Lombardini, C., Porhaly, J., & Spinelli, R. (2014). Soil Compaction and Recovery 528 after Mechanized Final Felling of Italian Coastal Pine Plantations. Croatian Journal of Forest 529 Engineering, 35(1), 63-71. 530 Klimešová, J., Martínková, J., Ottaviani, G., & Field, K. (2018). Belowground plant functional ecology: 531 Towards an integrated perspective. Functional Ecology, 32(9), 2115-2126. doi:10.1111/1365- 532 2435.13145 533 Kobe, R. K., Iyer, M., & Walters, M. B. (2010). Optimal partitioning theory revisited: Nonstructural 534 carbohydrates dominate root mass responses to nitrogen. Ecology, 91(1), 166-179. 535 doi:10.1890/09-0027.1 536 Kong, D., Wang, J., Wu, H., Valverde-Barrantes, O. J., Wang, R., Zeng, H., . . . Feng, Y. (2019). 537 Nonlinearity of root trait relationships and the root economics spectrum. Nature 538 Communications, 10(1), 2203. doi:10.1038/s41467-019-10245-6 539 Kopittke, P. M., Menzies, N. W., Wang, P., McKenna, B. A., & Lombi, E. (2019). Soil and the 540 intensification of agriculture for global food security. Environment International, 132, 105078. 541 doi:10.1016/j.envint.2019.105078 542 Kou, D., Yang, G., Li, F., Feng, X., Zhang, D., Mao, C., . . . Yang, Y. (2020). Progressive nitrogen 543 limitation across the Tibetan alpine permafrost region. Nature Communications, 11(1), 3331. 544 doi:10.1038/s41467-020-17169-6 545 LeBauer, D. S., & Treseder, K. K. (2008). Nitrogen limitation of net primary productivity in terrestrial 546 ecosystems is globally distributed. Ecology, 89(2), 371-379. doi:https://doi.org/10.1890/06- 547 2057.1 548 Leblans, N. I. W., Sigurdsson, B. D., Aerts, R., Vicca, S., Magnússon, B., & Janssens, I. A. (2017). 549 Icelandic grasslands as long-term C sinks under elevated organic N inputs. Biogeochemistry, 550 134(3), 279-299. doi:10.1007/s10533-017-0362-5 551 Leblans, N. I. W., Sigurdsson, B. D., Roefs, P., Thuys, R., Magnússon, B., & Janssens, I. A. (2014). 552 Effects of seabird nitrogen input on biomass and carbon accumulation after 50 years of primary 553 succession on a young volcanic island, Surtsey. Biogeosciences, 11(22), 6237-6250. 554 doi:10.5194/bg-11-6237-2014 555 Leblans, N. I. W., Sigurdsson, B. D., Vicca, S., Fu, Y., Penuelas, J., & Janssens, I. A. (2017). Phenological 556 responses of Icelandic subarctic grasslands to short-term and long-term natural soil warming. 557 Global Change Biology, 23(11), 4932-4945. doi:10.1111/gcb.13749 558 Leppälammi-Kujansuu, J., Ostonen, I., Strömgren, M., Nilsson, L. O., Kleja, D. B., Sah, S. P., & 559 Helmisaari, H. S. (2012). Effects of long-term temperature and nutrient manipulation on 560 Norway spruce fine roots and mycelia production. Plant and Soil, 366(1-2), 287-303. 561 doi:10.1007/s11104-012-1431-0 562 Lewis, J. D., Lucash, M., Olszyk, D., & Tingey, D. T. (2011). Seasonal patterns of photosynthesis in 563 Douglas fir seedlings during the third and fourth year of exposure to elevated CO 2 and 564 temperature. Plant, Cell & Environment, 24(5), 539-548. doi:DOI: 10.1046/j.1365- 565 3040.2001.00700.x 566 Liu, H., Mi, Z., Lin, L., Wang, Y., Zhang, Z., Zhang, F., . . . He, J. S. (2018). Shifting plant species 567 composition in response to climate change stabilizes grassland primary production. roceedings 568 of the National Academy of Sciences of the United States of America, 115(16), 4051-4056. 569 doi:10.1073/pnas.1700299114 570 Liu, L., Zhang, X., Xu, W., Liu, X., Lu, X., Wei, J., . . . Wong, A. Y. H. (2020). Reviewing global estimates 571 of surface reactive nitrogen concentration and deposition using satellite retrievals. Atmospheric 572 Chemistry and Physics, 20(14), 8641-8658. doi:10.5194/acp-20-8641-2020 573 Liu, Q., Piao, S., Janssens, I. A., Fu, Y., Peng, S., Lian, X., . . . Wang, T. (2018). Extension of the growing 574 season increases vegetation exposure to frost. Nature Communications, 9(1), 426. 575 doi:10.1038/s41467-017-02690-y 576 Liu, Y., Liu, S., Wan, S., Wang, J., Wang, H., & Liu, K. (2017). Effects of experimental throughfall 577 reduction and soil warming on fine root biomass and its decomposition in a warm temperate 578 oak forest. Science of the Total Environment, 574, 1448-1455. 579 doi:10.1016/j.scitotenv.2016.08.116 580 Lõhmus, K., Oja, T., & Lasn, R. (1989). Specific root area: A soil characteristic. Plant and Soil, 119(2), 581 245-249. doi:https://doi.org/10.1007/BF02370415 582 Luo, Y., Wan, S., Hui, D., & Wallace, L. L. (2001). Acclimatization of soil respiration to warming in a 583 tall grass prairie. Nature, 413(6856), 622-625. doi:10.1038/35098065 584 Ma, C., Zhang, W., Wu, M., Xue, Y., Ma, L., & Zhou, J. (2012). Effect of aboveground intervention on 585 fine root mass, production, and turnover rate in a Chinese cork oak (Quercus variabilis Blume) 586 forest. Plant and Soil, 368(1-2), 201-214. doi:10.1007/s11104-012-1512-0 587 Ma, H., L. Mo, T. W. Crowther, D. S. Maynard, J. van den Hoogen, B. D. Stocker, C. Terrer, and C. M. 588 Zohner. 2021. The global distribution and environmental drivers of aboveground versus 589 belowground plant biomass. Nature Ecology & Evolution, 5, 1110–1122. 590 https://doi.org/10.1038/s41559-021-01485-1 591 Ma, Z., Liu, H., Mi, Z., Zhang, Z., Wang, Y., Xu, W., . . . He, J. S. (2017). Climate warming reduces the 592 temporal stability of plant community biomass production. Nature Communications, 8, 15378. 593 doi:10.1038/ncomms15378 594 Maestre, F. T., Escolar, C., de Guevara, M. L., Quero, J. L., Lazaro, R., Delgado-Baquerizo, M., . . . 595 Gallardo, A. (2013). Changes in biocrust cover drive carbon cycle responses to climate change 596 in drylands. Global Change Biology, 19(12), 3835-3847. doi:10.1111/gcb.12306 597 Marañón-Jiménez, S., Peñuelas, J., Richter, A., Sigurdsson, B. D., Fuchslueger, L., Leblans, N. I. W., & 598 Janssens, I. A. (2019). Coupled carbon and nitrogen losses in response to seven years of chronic 599 warming in subarctic soils. Soil Biology and Biochemistry. doi:10.1016/j.soilbio.2019.03.028 600 Marañón-Jiménez, S., Soong, J. L., Leblans, N. I. W., Sigurdsson, B. D., Peñuelas, J., Richter, A., . . . 601 Janssens, I. A. (2018). Geothermally warmed soils reveal persistent increases in the respiratory 602 costs of soil microbes contributing to substantial C losses. Biogeochemistry, 138(3), 245-260. 603 doi:10.1007/s10533-018-0443-0 604 Mautner, M. (1999). Formation, chemistry and fertility of extraterrestrial soils: Cohesion, water 605 adsorption and surface area of carbonaceous chondrite. Prebiotic and space resource 606 applications. Icarus, 137(1), 178-195. doi:10.1006/icar.1998.6028 607 McCormack, M. L., Dickie, I. A., Eissenstat, D. M., Fahey, T. J., Fernandez, C. W., Guo, D., . . . Zadworny, 608 M. (2015). Redefining fine roots improves understanding of below-ground contributions to 609 terrestrial biosphere processes. New Phytologist, 207(3), 505-518. doi:10.1111/nph.13363 610 Melillo, J. M., Butler, S., Johnson, J., Mohan, J., Steudler, P., Lux, H., . . . Tang, J. (2011). Soil warming, 611 carbon-nitrogen interactions, and forest carbon budgets. Proceedings of the National Academy 612 of Sciences of the United States of America, 108(23), 9508-9512. doi:10.1073/pnas.1018189108 613 Melillo, J. M., Steudler, P. A., Aber, J. D., Newkirk, K., Lux, H., Bowles, F. P., . . . Morrisseau, S. (2002). 614 Soil warming and carbon-cycle feedbacks to the climate system. Science, 298(5601), 2173-2176. 615 doi:10.1126/science.1074153 616 Melillo, J. M., Frey, S. D., DeAngelis, K. M., Werner, W. J., Bernard, M. J., Bowles, F. P., . . . Grandy, 617 A. S. (2017). Long-term pattern and magnitude of soil carbon feedback to the climate system in 618 a warming world. Science, 358(6359), 101-104. doi:10.1126/science.aan2874 619 Michielsen L 2014. Plant communities and global change: adaptation by changes in present species 620 composition or adaptation in plant traits. A case study in Iceland. M.Sc. thesis. University of 621 Antwerp, Antwerp, Belgium. 53 p 622 Mokany, K., R. J. Raison, and A. S. Prokushkin. 2006. Critical analysis of root:shoot ratios in terrestrial 623 biomes. Global Change Biology, 12:84-96. https://doi.org/10.1111/j.1365-2486.2005.001043.x 624 Myrstener, M., Rocher-Ros, G., Burrows, R. M., Bergstrom, A. K., Giesler, R., & Sponseller, R. A. 625 (2018). Persistent nitrogen limitation of stream biofilm communities along climate gradients in 626 the Arctic. Global Change Biology, 24(8), 3680-3691. doi:10.1111/gcb.14117 627 Nijs, I., Kockelbergh, F., Teughels, H., Blum, H., Hendrey, G., & Impens, I. (1996). Free Air Temperature 628 Increase (FATI): a new tool to study global warming effects on plants in the field. Plant, Cell & 629 Environment, 19(4), 495-502. doi:10.1111/j.1365-3040.1996.tb00343.x 630 Noyce, G. L., Kirwan, M. L., Rich, R. L., & Megonigal, J. P. (2019). Asynchronous nitrogen supply and 631 demand produce nonlinear plant allocation responses to warming and elevated CO 2. roceedings 632 of the National Academy of Sciences of the United States of America, 116(43), 21623-21628. 633 doi:10.1073/pnas.1904990116 634 Onoz, B., & Oguz, B. (2003). Assessment of Outliers in Statistical Data Analysis. In: Harmancioglu 635 N.B., Ozkul S.D., Fistikoglu O., Geerders P. (eds) Integrated Technologies for Environmental 636 Monitoring and Information Production. Nato Science Series (Series: IV: Earth and 637 Environmental Sciences), 23. doi:https://doi.org/10.1007/978-94-010-0231-8_13 638 Ottaviani, G., R. Molina-Venegas, T. Charles-Dominique, S. Chelli, G. Campetella, R. Canullo, and J. 639 Klimesova. 2020. The Neglected Belowground Dimension of Plant Dominance. Trends Ecology 640 and Evolution, 35:763-766. https://doi.org/10.1016/j.tree.2020.06.006 641 Peek, M. S., Leffler, A. J., Hipps, L., Ivans, S., Ryel, R. J., & Caldwell, M. M. (2006). Root turnover and 642 relocation in the soil profile in response to seasonal soil water variation in a natural stand of 643 Utah juniper (Juniperus osteosperma). Tree Physiology, 26, 1469–1476. 644 Penuelas, J., Poulter, B., Sardans, J., Ciais, P., van der Velde, M., Bopp, L., . . . Janssens, I. A. (2013). 645 Human-induced nitrogen-phosphorus imbalances alter natural and managed ecosystems across 646 the globe. Nature Communications, 4, 2934. doi:10.1038/ncomms3934. 647 Poorter, H., K. J. Niklas, P. B. Reich, J. Oleksyn, P. Poot, and L. Mommer. 2012. Biomass allocation to 648 leaves, stems and roots: meta-analyses of interspecific variation and environmental control. New 649 Phytologist, 193, 30-50. https://doi.org/10.1111/j.1469-8137.2011.03952.x 650 Price, M. V., and N. M. Waser. 1998. Effects of experimental warming on plant reproductive phenology 651 in a subalpine meadow. Ecology, 79, 1261-1271. https://doi.org/10.1890/0012- 652 9658(1998)079[1261:EOEWOP]2.0.CO;2 653 Pries, C. E. H., Castanha, C., Porras, R. C., & Torn, M. S. (2017). The whole-soil carbon flux in response 654 to warming. Science, 355(6332), 1420-1422. doi:10.1126/science.aal1319 655 Qi, Y., Wei, W., Chen, C., & Chen, L. (2019). Plant root-shoot biomass allocation over diverse biomes: 656 A global synthesis. Global Ecology and Conservation, 18, e00606. 657 doi:10.1016/j.gecco.2019.e00606 658 Quinn, G. P., & Keough, M. J. (2009). Experimental Design and Data Analysis for Biologists. Cambridge: 659 Cambridge University Press. 660 Radujkovic, D., Verbruggen, E., Sigurdsson, B. D., Leblans, N. I. W., Janssens, I. A., Vicca, S., & Weedon, 661 J. T. (2018). Prolonged exposure does not increase soil microbial community compositional 662 response to warming along geothermal gradients. FEMS Microbiology Ecology, 94(2), fix174. 663 doi:10.1093/femsec/fix174 664 Raich, J. W., & Schlesinger, W. H. (1992). The global carbon dioxide flux in soil respiration and its 665 relationship to vegetation and climate. Tellus, 44(2), 81-99. doi:10.1034/j.1600-0889.1992.t01- 666 1-00001.x 667 Rawls WJ, Pachepsky YA, Ritchie JC, Sobecki TM, Bloodworth H. 2003. Effect of soil organic carbon 668 on soil water retention. Geoderma 116(1-2), 61-76. 669 Rustad, L. E. (2008). The response of terrestrial ecosystems to global climate change: towards an 670 integrated approach. Science of the Total Environment, 404(2-3), 222-235. 671 doi:10.1016/j.scitotenv.2008.04.050 672 Rygiewicz, P. T., & Andersen, C. P. (1994). Mycorrhizae alter quality and quantity of carbon allocated 673 below ground. Nature(369), 58–60. doi:https://doi.org/10.1038/369058a0 674 Salazar, A., K. Rousk, I. S. Jónsdóttir, J.-P. Bellenger, and Ó. S. Andrésson. 2020. Faster nitrogen cycling 675 and more fungal and root biomass in cold ecosystems under experimental warming: a meta- 676 analysis. Ecology 101:e02938. 677 Schaeffer, S. M., Sharp, E., Schimel, J. P., & Welker, J. M. (2013). Soil-plant N processes in a High 678 Arctic ecosystem, NW Greenland are altered by long-term experimental warming and higher 679 rainfall. Global Change Biology, 19(11), 3529-3539. doi:10.1111/gcb.12318 680 Schmidt, M. W., Torn, M. S., Abiven, S., Dittmar, T., Guggenberger, G., Janssens, I. A., . . . Trumbore, 681 S. E. (2011). Persistence of soil organic matter as an ecosystem property. Nature, 478(7367), 682 49-56. doi:10.1038/nature10386 683 Schwieger, S., Kreyling, J., Milbau, A., & Blume-Werry, G. (2018). Autumnal warming does not change 684 root phenology in two contrasting vegetation types of subarctic tundra. Plant and Soil, 424(1- 685 2), 145-156. doi:10.1007/s11104-017-3343-5 686 See, C. R., Luke McCormack, M., Hobbie, S. E., Flores-Moreno, H., Silver, W. L., & Kennedy, P. G. 687 (2019). Global patterns in fine root decomposition: climate, chemistry, mycorrhizal association 688 and woodiness. Ecology Letters, 22(6), 946-953. doi:10.1111/ele.13248 689 Semchenko, M., Leff, J. W., Lozano, Y. M., Saar, S., Davison, J., Wilkinson, A., . . . Bardgett, R. D. 690 (2018). Fungal diversity regulates plant-soil feedbacks in temperate grassland. Science 691 Advances, 4(11), eaau4578. doi:10.1126/sciadv.aau4578 692 Sherry, R. A., Weng, E., Arnone Iii, J. A., Johnson, D. W., Schimel, D. S., Verburg, P. S., . . . Luo, Y. 693 (2008). Lagged effects of experimental warming and doubled precipitation on annual and 694 seasonal aboveground biomass production in a tallgrass prairie. Global Change Biology, 14(12), 695 2923-2936. doi:10.1111/j.1365-2486.2008.01703.x 696 Sigurdsson, B. D., Leblans, N. I. W., Dauwe, S., Guðmundsdóttir, E., Gundersen, P., Gunnarsdóttir, G. 697 E., . . . Janssens, I. (2016). Geothermal ecosystems as natural climate change experiments: The 698 ForHot research site in Iceland as a case study. Icelandic Agricultural Sciences, 29, 53-71. 699 doi:10.16886/ias.2016.05 700 Sistla, S. A., Moore, J. C., Simpson, R. T., Gough, L., Shaver, G. R., & Schimel, J. P. (2013). Long-term 701 warming restructures Arctic tundra without changing net soil carbon storage. Nature, 497(7451), 702 615-618. doi:10.1038/nature12129 703 Slessarev EW, Nuccio EE, McFarlane KJ, Ramon CE, Saha M, Firestone MK, Pett-Ridge J. (2020). 704 Quantifying the effects of switchgrass (Panicum virgatum) on deep organic C stocks using 705 natural abundance 14C in three marginal soils. GCB Bioenergy 12(10): 834-847. 706 Song, J., Wan, S., Piao, S., Knapp, A. K., Classen, A. T., Vicca, S., . . . Zheng, M. (2019). A meta-analysis 707 of 1,119 manipulative experiments on terrestrial carbon-cycling responses to global change. 708 Nature Ecology & Evolution. doi:10.1038/s41559-019-0958-3 709 Thomas, R. Q., Zaehle, S., Templer, P. H., & Goodale, C. L. (2013). Global patterns of nitrogen limitation: 710 confronting two global biogeochemical models with observations. Global Change Biology, 711 19(10), 2986-2998. doi:10.1111/gcb.12281 712 Verbrigghe N, Leblans NIW, Sigurdsson BD, Vicca S, Fang C, Fuchslueger L, … Janssens, I. A. (2022). 713 Soil carbon loss in warmed subarctic grasslands is rapid and restricted to topsoil. Biogeosciences, 714 19(14), 3381-3393. 715 Ven, A., Verlinden, M. S., Verbruggen, E., & Vicca, S. (2019). Experimental evidence that phosphorus 716 fertilization and arbuscular mycorrhizal symbiosis can reduce the carbon cost of phosphorus 717 uptake. Functional Ecology. doi:10.1111/1365-2435.13452 718 Vicca, S., Luyssaert, S., Penuelas, J., Campioli, M., Chapin, F. S., 3rd, Ciais, P., . . . Janssens, I. A. (2012). 719 Fertile forests produce biomass more efficiently. Ecology Letters, 15(6), 520-526. 720 doi:10.1111/j.1461-0248.2012.01775.x 721 Walker, T. W. N., Janssens, I. A., Weedon, J. T., Sigurdsson, B. D., Richter, A., Penuelas, J., . . . 722 Verbruggen, E. (2020). A systemic overreaction to years versus decades of warming in a 723 subarctic grassland ecosystem. Nature Ecology & Evolution, 4(1), 101-108. 724 doi:10.1038/s41559-019-1055-3 725 Walker, T. W. N., Kaiser, C., Strasser, F., Herbold, C. W., Leblans, N. I. W., Woebken, D., . . . Richter, A. 726 (2018). Microbial temperature sensitivity and biomass change explain soil carbon loss with 727 warming. Nature Climate Change, 8(10), 885-889. doi:10.1038/s41558-018-0259-x 728 Wan, S., Hui, D., Wallace, L., & Luo, Y. (2005). Direct and indirect effects of experimental warming on 729 ecosystem carbon processes in a tallgrass prairie. Global Biogeochemical Cycles, 19(2), 730 GB2014. doi:10.1029/2004gb002315 731 Wang, N., B. Quesada, L. Xia, K. Butterbach-Bahl, C. L. Goodale, and R. Kiese. (2019). Effects of 732 climate warming on carbon fluxes in grasslands- A global meta-analysis. Ecology Letters, 733 25,1839-1851. 734 Wilson, J. B. (1988). A Review of Evidence on the Control of Shoot: Root Ratio, in Relation to Models. 735 Annals of Botany, 61(4), 433-449. doi:10.1093/oxfordjournals.aob.a087575 736 Xia, J., & Wan, S. (2013). Independent effects of warming and nitrogen addition on plant phenology in 737 the Inner Mongolian steppe. Annals of Botany, 111(6), 1207-1217. doi:10.1093/aob/mct079 738 Xiong, D., Yang, Z., Chen, G., Liu, X., Lin, W., Huang, J., . . . Yang, Y. (2018). Interactive effects of 739 warming and nitrogen addition on fine root dynamics of a young subtropical plantation. Soil 740 Biology and Biochemistry, 123, 180-189. doi:10.1016/j.soilbio.2018.05.009 741 Xu, X., Niu, S., Sherry, R. A., Zhou, X., Zhou, J., & Luo, Y. (2012). Interannual variability in responses 742 of belowground net primary productivity (NPP) and NPP partitioning to long-term warming and 743 clipping in a tallgrass prairie. Global Change Biology, 18(5), 1648-1656. doi:10.1111/j.1365- 744 2486.2012.02651.x 745 Table 1 Mean annual soil temperature from July 2017 to July 2018 in situ soil warming. Different 746 letters indicate significant difference among warming treatments at p < 0.05 (one-way ANOVAs 747 and Tukey Kramer post hoc tests). Transect +0 ºC +3.3 ºC +7.9 ºC 1 5.5 8.9 10.9 2 6.1 8.0 11.2 3 6.3 9.2 15.7 4 4.2 9.4 14.3 5 7.1 10.2 16.8 Mean annual soil T 5.8±1.0c 9.1±0.7b 13.8±2.4a (ºC) 748 749 Table 2 Pearson correlation coefficients (r values) between plant traits and soil temperature, bulk density and nutrients. Bulk Volumetric Gravimetric Nutrients concentration Nutrients pool Plant traits Temperature + 2+ 2+ density water content water content N P K Ca Mg N P K+ Ca2+ Mg2+ Shoot production -0.67* -0.45 0.81** 0.63* 0.96** 0.75* 0.03 -0.67* -0.43 0.79** 0.41 0.79** -0.27 -0.14 Fine root production -0.77** -0.69** 0.69** 0.81** 0.71** 0.75** -0.20 -0.65* -0.45 0.55 0.30 0.30 -0.28 -0.22 Standing root biomass -0.55* -0.44 0.13 0.42 0.50++ 0.54* -0.09 -0.45 -0.14 0.30 -0.07 -0.23 -0.56++ -0.41 Moss-Vascular plant ratio 0.67* 0.51 -0.74** -0.61* -0.76** -0.90** 0.07 0.67* 0.57++ -0.69* -0.55++ -0.53 -0.05 -0.06 Root turnover 0.13 0.11 -0.37 -0.22 -.66* -0.21 -0.17 0.10 0.36 -0.69* -0.35 -0.56++ -0.17 0.005 Root-shoot ration -0.28 -0.43 0.46 0.46 0.29 0.37 -0.04 -0.24 -0.25 0.22 0.38 0.44 0.36 0.27 Specific root area 0.71** 0.74** -0.51++ -0.81** -0.77** -0.60* -0.16 0.63* 0.50++ -0.59* -0.07 -0.50 0.44 0.45 Fine root N -0.12 -0.35 0.17 0.36 0.40 0.41 -0.34 -0.60* -0.58* 0.32 -0.41 -0.67* -0.64* -0.63++ ++ 750 p < 0.10, * p < 0.05, ** p < 0.01. 751 Table 3 Linear stepwise regression models between vascular plant production and other 752 plant traits, soil temperature, bulk density and nutrients based on Pearson’s correlation. 753 N: soil N concentration; SRA: specific root area; SWV: soil volumetric water content. Model F p R2 Shoot production = 2.943N-0.806 21.287 <0.001 0.628 Fine root production = -0.363SRA + 54.054 31.497 <0.001 0.741 Fine root production = -0.292SRA + 0.929SWV + 11.446 26.723 <0.001 0.842 754 755 Figure captions: 756 Fig. 1 Fine root production and shoot biomass in 2018 under +0 ºC, +3.3 ºC and +7.9 757 ºC treatments. W effect: warming effect. p < 0.05 means significant soil warming effect. 758 Different lowercase letters represent significant difference at p < 0.05. 759 760 Fig. 2 the biomass ratio of moss and vascular plant shoot part (Moss/Grass) (E), 761 standing root biomass (B), root turnover (C), root-shoot ratio (D), specific root area 762 (SRA) (E) and fine root N concentration (E) in 2018 under +0 ºC, +3.3 ºC and +7.9 ºC 763 treatments. W effect: warming effect. p < 0.05 means significant soil warming effect; 764 NS means no significant soil warming effect. Red shaded areas indicate a 95% 765 confidence interval. Different lowercase letters represent significant difference at p < 766 0.05. 767 768 Fig. 3 Mean soil water content during growing seasons from 2013-2017 and soil bulk 769 density in 2018 under +0 ºC, +3.3 ºC and +7.9 ºC treatments. Soil gravimetric water 770 content was calculated as soil volumetric water divided by soil bulk density. W effect: 771 warming effect. p < 0.05 means significant soil warming effect; NS means no significant 772 soil warming effect. Red shaded areas indicate a 95% confidence interval. Different 773 lowercase letters represent significant difference at p < 0.05. 774 775 Fig. 4 Soil nutrient concentrations (A-E) and pools (F-L) under +0 ºC, +3.3 ºC and +7.9 776 ºC treatments. Soil N data were collected in July 2018 and other soil nutrients data were 777 collected in July 2014. W effect: warming effect. p < 0.05 means significant soil 778 warming effect; NS means no significant soil warming effect. Red shaded areas indicate 779 a 95% confidence interval. Different lowercase letters represent significant difference 780 at p < 0.05. 781 782 Fig. 5 Relationships of above- and belowground biomass and production with soil 783 temperature, soil bulk density, Moss/Grass, soil N concentration. Significance is at p < 784 0.05. 785 786 Fig. 6 Structure equation modelling (SEM) with variables (boxes) and potential causal 787 relationships (arrows) for fine root production (χ2 = 1.379, df = 3, p = 0.71 > 0.05, CFI 788 = 1.00 > 0.9). BD: bulk soil density; TN: soil N concentration; T: measured mean annual 789 soil temperature; Moss/Vascular plant: Moss-vascular plant ratio in above-ground 790 biomass; SRA: specific root area. Single headed arrows represent the hypothesized 791 direction of causation. Numbers next to single headed arrows are standardized path 792 coefficients, which indicate the effect size of the relationship. Black and red arrows 793 indicate positive and negative relationship. The proportion of variance explained 794 appears above each response variables in the model. 795 796 Fig. 7 Structure equation modelling (SEM) with variables (boxes) and potential causal 797 relationships (arrows) for shoot production (χ2 = 6.07, df = 3, p = 0.108 > 0.05, CFI = 798 0.94 > 0.9). BD: bulk soil density; TN: soil N concentration; T: measured mean annual 799 soil temperature; Moss/Vascular plant: Moss-vascular plant ratio in above-ground 800 biomass. Single headed arrows represent the hypothesized direction of causation. 801 Numbers next to single headed arrows are standardized path coefficients, which 802 indicate the effect size of the relationship. The double headed arrow represents the 803 covariance between related variables. Black and red arrows indicate positive and 804 negative relationship. The proportion of variance explained appears above each 805 response variables in the model. Solid line means significance and dash line means non- 806 significance. 807 808 Fig. 1 809 810 Fig. 2 811 812 Fig. 3 813 814 Fig. 4 815 816 Fig. 5 817 818 Fig. 6 819 820 Fig. 7 821 822 Appendix A 823 Table A1 Data amount of variables in analyzing difference among treatments. Standing Shoot Fine root Root Fine Treatment Root Moss/Grass T SWC RS BD N P K+ Ca2+ Mg2+ SRA production production turnover root N biomass +0 oC 3 5 4 4 5 4 5 5 3 5 5 3 3 3 3 5 +3.3 oC 4 5 5 5 5 4 5 5 4 5 4 4 4 4 4 3 o +7.9 C 4 5 5 5 5 4 5 5 4 5 5 4 4 4 4 5 824 825 Table A2 Data amount of variable combinations in analyzing relationship between biomass and 826 environment factors Variables T BD Moss/Grass TN SRA Shoot production 11 11 11 11 11 Standing root 15 15 12 14 14 biomass Fine root production 14 14 11 13 13 827 828 Fig. A1 Conceptual structure equation modelling (SEM) with variables (boxes) and 829 potential causal relationships (arrows) for shoot and fine root production. T: soil 830 temperature; N: soil nitrogen concentration; BD: soil bulk density; SRA: specific root 831 area. Due to high soil water content in this study site, soil water was not considered as 832 a limiting factor of plant growth and thus not included into SEM. 833 834 Fig. A2 Total fine roots area under +0 ºC, +3.3 ºC and +7.9 ºC treatments. Different 835 letters represent significant difference at p < 0.05. 836 837 Fig. A3 Effect of warming on moss biomass under +0 ºC, +3.3 ºC and +7.9 ºC 838 treatments. Different letters represent significant difference at p < 0.05. 839 840 Fig. A4 Soil nitrogen and phosphorus ratio under +0 ºC, +3.3 ºC and +7.9 ºC treatments. 841 Different letters represent significant difference at p < 0.05. 842 843 Fig. A5 Effect of warming on spring plant phenology. Photo was taken on 26 April in 844 2018. This photo showed earlier onset of vegetation with warming. 845 846 Fig. A6 Total soil cations pool under +0 ºC, +3.3 ºC and +7.9 ºC treatments. Different 847 letters represent significant difference at p < 0.05. 848 Fig. A1 849 850 Fig. A2 851 852 Fig. A3 853 854 Fig. A4 855 856 Fig. A5 857 858 Fig. A6 859