FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964646550> ?p ?o ?g. }

• W2964646550 endingPage "2041" @default.
• W2964646550 startingPage "2030" @default.
• W2964646550 abstract "根系性状的种内变异反映了植物对变化环境的适应策略，是植物在变化环境中生存和生长的关键因素。降水变化是影响草地生态系统结构和功能重要的全球变化因子之一。然而，很少有研究探讨在野外条件下草原植物根系性状对降水变化的响应。本研究利用在内蒙古温带草原开展的野外长期（10年）增加降水控制实验和沿着从东到西的自然降水梯度(412.3–144.2 mm)设置的长约700 km的样带，分别选取了15种常见草本物种和2种区域广布种为研究对象，对其根系形态性状、化学性状和解剖结构进行测定，以期探讨草原植物对降水变化的适应策略。我们的研究结果发现：1）内蒙古典型草原植物的大部分根系性状对增加降水无显著响应，少数几个物种通过改变某几个根性状来适应降水的改变。沿着自然降水梯度，只有当降水量改变到一定程度之后，物种的表型性状才会有适应性调整。这些结果表明，内蒙古典型草原植物根系性状对降水变化响应的敏感性较小，可以适应一定范围的降水变化。2）在增加降水实验中，羊草的根系性状对增加降水无显著响应，而增加降水使冷蒿的比根长和比表面积降低而组织密度升高。沿着自然降水梯度，羊草在降水量低的样点有较高的比根长、比表面积和根N浓度，有较低的组织密度和C：N比，而冷蒿在降水量低的样点有较高的根系直径，和较低的比根长和比表面积。此结果表明，不同的物种通过多样的根系性状调整策略来适应降水的变化。 综上所述，内蒙古温带草原植物的根系性状对降水变化的响应敏感性较小，且不同植物种通过不同的表型调整策略来适应降水的变化。本研究为理解和预测温带草原植物对未来气候变化的响应提供了重要的理论基础。 A free Plain Language Summary can be found within the Supporting Information of this article. Root traits can strongly influence growth, survival and response of plants to climate change due to their essential roles in water and nutrient acquisition and transportation (Fort et al., 2015; Grime, 1974; Iversen et al., 2017; Kong et al., 2014; Lambers, Shane, Cramer, Pearse, & Veneklaas, 2006; Larson & Funk, 2016; Mueller et al., 2018; Warren et al., 2015). The responses of absorptive root traits to changing environments can reflect the adaptive strategies of resource utilization as well as plant performance under future climate change (Bardgett, Mommer, & De Vries, 2014; Freschet, Swart, & Cornelissen, 2015; Hodge, 2004; Warren et al., 2015). However, few studies have focused on how absorptive root traits respond to climate change (Iversen et al., 2017; Larson & Funk, 2016; Zadworny, McCormack, Mucha, Reich, & Oleksyn, 2016). Filling up this knowledge gap is of importance in semi-arid grasslands as most of the biomass is allocated into below-ground (Jackson et al., 1996; Mokany, Raison, & Prokushkin, 2006). In addition, water is the most important factor that determines growth and survival of plants and limits production and stability of terrestrial ecosystem in semi-arid grasslands (Bai et al., 2010; Yang et al., 2011). It has been established that shift in precipitation is one of the most pertinent global change factors to regulate grassland functions (Christensen, Coughenour, Ellis, & Chen, 2004; Liu et al., 2013). Thus, characterizing the response of absorption root traits to changing water availability is crucial for understanding the resource utilization strategies and performance of plant species in grasslands under future climate change. Plants can change their capacity for resource acquisition and/or alter the needs for the resources in response to changing resource availability. However, the direction and magnitude of the response of root traits to the changes in resource availability have been under debate (Bloom, Chapin, & Mooney, 1985; Chapin, 1980; Freschet, Violle, Bourget, Scherer-Lorenzen, & Fort, 2018; Reich, 2018). A hypothesis referred to the resource economics hypothesis has been proposed. According to the resource economic spectrum, plant species generally tend to have acquisitive trait syndromes (e.g. higher specific leaf area, specific root length [SRL] and tissue nitrogen content) at resource abundant environments for maximal acquisition of resource and fast growth. In contrast, plant species evolved conservative trait syndromes (e.g. higher tissue density, larger root diameter, lower tissue nitrogen content) to conserve resource, thus ensuring the survival under low-resource conditions (Larson & Funk, 2016; Reich & Cornelissen, 2014; Weemstra, Mommer, et al., 2016a; Wright & Westoby, 1999). It has been predicted that a similar intraspecific variation in root traits in response to changing water availability may also occur along the precipitation gradients. On the other hand, a contrasting hypothesis has been proposed following the functional equilibrium hypothesis (Brouwer, 1962). The functional equilibrium hypothesis suggests that plants would allocate more biomass to the organs involved in absorption of the limiting resource. Similarly, plants would be more acquisitive under conditions of lower resource availability and more conservative when grown in higher resource availability environments (Holdaway, Richardson, Dickie, Peltzer, & Coomes, 2011; Prieto et al., 2015; Weemstra, Sterck, et al., 2016b). Several studies exploring the intraspecific responses of root morphological and chemical traits to changing water availability have revealed that both of the hypotheses are tenable (Larson & Funk, 2016; Nguyen et al., 2016; Padilla, Miranda, Jorquera, & Pugnaire, 2009; Ryalls, Moore, Johnson, Connor, & Hiltpold, 2018). For example, Ryalls et al. (2018) found that the SRL of a meadow grass (Microlaena stipoides) decreased and diameter and tissue density increased under deluge. These results indicate that plants become less acquisitive under water addition, thus supporting the functional equilibrium hypothesis. By contrast, Isaac et al. (2017) reported that Coffea arabica grown in cool and dry environments had higher root diameters, lower SRL and root nitrogen contents than plants grown in cool and wet environments. This result suggests that plants would be more conservative under drier conditions, and is consistent with the resource economics hypothesis. Furthermore, a synthesis of global field trials revealed that the SRL decreased for grass and forbs, but increased for woody species under drought conditions (Zhou, Zhou, et al., 2018). Larson and Funk (2016) reported that diverse directions and extents of responses in root traits to watering treatments were detected both in seedlings of woody and herbaceous species. However, Nguyen et al. (2016) found no phenotypic variation in the examined root traits among water treatments. Therefore, the mechanisms underlying the diverse response patterns warrant further exploration. The results on responses of root traits to water treatments mainly came from the greenhouse experiments (Larson & Funk, 2016; Nguyen et al., 2016; Padilla et al., 2009; Ryalls et al., 2018). However, the interactions between plants and environments are more complex and variable in field conditions than in controlled conditions, thus leading to substantial differences in morphological traits between controlled environments and under field conditions (Poorter et al., 2016). Therefore, the responses of root traits to changes in precipitation warrant detailed investigation in field conditions. Though many manipulated field experiments have been conducted to simulate precipitation change in grasslands (Bai et al., 2010; Niu et al., 2008; Wilcox et al., 2017; Xu et al., 2015; Zhang et al., 2017), few studies have specifically focused on the effects of changing precipitation on root traits. Changes in root anatomical traits such as stele diameter, cortex thickness and vessel diameter can also play important roles in water absorption under variable environments. For example, an increase in cortex thickness can provide more space for mycorrhizal symbiosis, thus enhancing the absorption efficiency (Guo et al., 2008; Kong et al., 2016). By contrast, a higher stele/root diameter ratio can increase the efficiency of water transportation (Kong et al., 2016). Further, vessels with larger diameters have been shown to be associated with water exploitation strategies (Pineda-Garcia, Paz, Meinzer, & Angeles, 2016). Nevertheless, few studies have integrated the morphological, chemical and anatomical root traits at species level to explore the adaptive strategies of plants in response to changing precipitation. In theory, plasticity of root traits can mainly be controlled by environmental and phylogenetic cues (Nicotra et al., 2010; Valladares, Gianoli, & Gomez, 2007). Previous studies suggested that the majority of root traits was phylogenetic conserved (Chen, Zeng, Eissenstat, & Guo, 2013; Kong et al., 2014; Zhou, Bai, Zhang, & Zhang, 2018). The phylogenetic constrains on root traits may strongly limit the plasticity of root traits in response to short-term changes in precipitation (manipulated experiments). At regional scales, the intraspecific variation in root traits along the environmental gradients integrates the response of plants to environmental change over time. This can provide comprehensive information to predict root response to future climate change compared with short-term manipulation experiments (Laughlin & Messier, 2015; Steppan, Phillips, & Houle, 2002). However, to the best of our knowledge, no studies have compared the responses of root traits to precipitation change between manipulation experiments at a local scale and along precipitation gradients at a regional scale. We hypothesize that the plasticity of root traits in response to short-term precipitation change (manipulated precipitation change experiments) may differ from that to long-term precipitation change (along the precipitation gradients) when phylogenetic constrains on root traits are considered. Studies on patterns of interspecific variation of absorption root traits both at local and regional scales have revealed multi-dimensional trade-off and diverse resource strategies of root traits among species and across environmental gradients (Chen et al., 2013; de la Riva et al., 2017; Kong et al., 2014; Roybal & Butterfield, 2018; Zhou, Bai, et al., 2018). These results suggest that different species can optimize different root traits to acquire the limited resources more efficiently. Thus, our second hypothesis is that the response of root traits to precipitation change is species-specific and trait-dependent both in the simulated increased precipitation experiment at local scale and along the precipitation gradients. Moreover, this may be one of the explanations to account for diverse patterns of variation in response to change water availability among different studies. To test our hypotheses, a 10-year field manipulated experiment to simulate precipitation change was conducted in a temperate grassland and a 700-km regional scale transect along precipitation gradients was established in northern China. As precipitation in summers has been projected to increase in the study area (Sun & Ding, 2009), water addition was used as a treatment in our manipulated experiments. We determined the morphological (average diameter [AD], SRL, root tissue density [RTD], specific root area [SRA]), chemical (root carbon content, root nitrogen content, root C/N ratio) and anatomical traits (stele diameter, cortex thickness, stele/ root diameter ratio, stele diameter/ cortex thickness, vessel density, vessel number, average vessel diameter) of the first two-order roots for the 15 herbaceous species in the manipulated experiment. We also evaluated these traits for the two regionally common species (grass Leymus chinensis, forb Artemisia frigida) along the precipitation gradients. The manipulated precipitation change experiment was conducted in a temperate steppe at the Duolun Restoration Ecology Experimentation and Demonstration Station, Institute of Botany, the Chinese Academy of Sciences, in the Duolun County, Inner Mongolia Autonomous Region, China (DREEDS, 116°17′E, 42°02′N; 1,324 m a.s.l.). The experimental site is distinguished by a semi-arid continental monsoon climate, with the mean annual temperature (MAT) of 2.1°C and mean annual precipitation (MAP) of 385.5 mm, respectively, with c. 80% total precipitation falling from June to September. Soil type is Calcis-orthic Aridisol. The soil contains 62.7% sand, 20.3% silt and 17.0% clay. Mean soil bulk density is 1.31 g/cm3, and pH is c. 7.0. The temperate steppe was dominated by perennial herbaceous species, including Stipa krylovii, A. frigida, Potentilla acaulis, Cleistogenes squarrosa, Allium bidentatum and Agropyron cristatum (Niu et al., 2008). Eight 15 m × 10 m plots were established with a distance of 1 m between each plot. Four plots were randomly assigned to water addition treatments; the other four plots were ambient precipitation treatment. Water of 15 mm was added weekly using six sprinklers in each plots in July and August from 2005 to 2015. A total of 120 mm water that was c. 30% of local annual precipitation was added in each year to simulate the projected change in precipitation (Sun & Ding, 2009). We selected 15 herbaceous species in the community for measuring their root traits. These species represent a broad range of taxa of the most common species in the temperate steppe, including eight dicotyledons and seven monocotyledons. Details of these species are listed in Table S1. Soil volumetric water content and inorganic N (NH4+–N, NO3−–N) concentrations were measured under conditions of control and water addition. Soil volumetric water content (0–10 cm) was measured with a portable device (Diviner 2000; Sentek Pty Ltd) at a 5-day interval (6 times per month) in July and August 2014. In each plot, three soil cores at 0–10 cm soil were taken and then mixed to form one composite sample in Mid-August of 2014. The soil samples were passed through a 2.0 mm sieve and homogenized. Fresh soil samples (10.0 g) were extracted with 50 ml of 2 M KCl solution and then analysed for inorganic N (NH4+–N, NO3−–N) concentrations using a continuous-flow ion auto-analyzer (Scalar SAN plus segmented flow analyzer). At least 10 mature plants of each species in each plot that were at a distance of c. 1 m with each other were selected randomly and excavated in early August 2014. Previous studies have reported that over 80% of the roots occurred in the top 30 cm soil in temperate grassland in north China (Ma, Yang, He, Zeng, & Fang, 2008). Thus, surface soil (at c. 30 cm depth) at the distance of 10–20 cm from the base of the plants was carefully excavated to obtain intact root system, and some deep rooted species were excavated to c. 50 cm depth. Soils adhered to the roots were carefully removed. Five intact root systems were immediately put into polyethylene terephthalate bottles filled with Formalin-Aceto-Alcohol (FAA) solution (90 ml of 50% ethanol, 5 ml of 100% acetic and 5 ml of 37% methanol) for later anatomical measurements. The remaining samples were put in plastic bags and kept in iceboxes, then transported to the laboratory within 4 hr and frozen at −20°C for subsequent morphological and chemical analyses. More than five intact root branches for each plant (five plants for each species were mixed to one sample in each plot) were used to determine morphological and chemical traits. To evaluate the intraspecific variation of root traits along the precipitation gradient, a 700-km regional scale transect was established in a temperate grassland in Inner Mongolia, northern China, ranging from 44°45ʹN to 46°42ʹN in latitude and 111°53ʹ to 119°17ʹE in longitude, with an elevation between 773 and 1,442 m (Figure S1). The transect was characterized by apparent precipitation gradients (Table S2). Along the transect, the MAP ranged from 412 mm in the east part to 144 mm in the west part of the transect, with minimum monthly precipitation in January (1.21–3.11 mm) and maximum monthly precipitation in July (128.13–38.75 mm), and c. 80% of the total precipitation occurred from June to September. The MAT ranged from −2.61 to 1.14°C, with minimal mean monthly temperatures in January (−24.89 to −20.64°C) and maximal mean monthly temperatures in July (17.34–21.46°C). The soil type was chestnut to brown soils along the east-west transect. The details of abiotic and biotic characteristics of the sampling sites along the transect are in Table S2. Along the precipitation gradient transect, 10 un-grazed grassland sites were selected along the east-west transect with a distance of c. 70-km between each site (Figure S1, Table S2). There was a decrease in annual precipitation across the sites from site 1 at east to the site 10 at west (Table S2). These sites belonged to three vegetation types (i.e. meadow steppe, typical steppe and desert steppe) and represented the dominant plant community types in the Inner Mongolia grassland and the Eurasian steppe region (Bai et al., 2008). The meadow steppe, located at the east part of the transect (sites 1–3), is dominated by Stipa baicalensis, Filifolium sibiricum and Carex pediformis. The typical steppe, located at the middle part of the transect (sites 4–7), is dominated by Stipa grandis, L. chinensis and C. squarrosa. The desert steppe, located at the west part of the transect (sites 8–10), is dominated by S. krylovii, Cleistogenes songorica, Stipa breviflora, Zygophyllum gobicum, Halogetong lomeratus and Agropyron desertorum (Cheng, Chu, Chen, & Bai, 2016). Following a simple vegetation survey (Table S3), two regionally common species, grass L. chinensis and forb A. frigida, were selected to explore the intraspecific variation of root traits along the precipitation gradients. At each site, at least 25 mature plants that were at a distance of c. 10 m with each other were selected randomly and excavated in August 2015. Similar with the manipulated precipitation change experiment, five intact root systems were immediately put into polyethylene terephthalate bottles filled with FAA solution for later anatomical measurements. More than five intact root branches from five plants were mixed into one subsample to determine the morphological and chemical traits at each site along the precipitation gradient transect. Four subsamples at each site were collected, leading to a total of 20 plants for each species at each site for determination of morphological and chemical traits. The root traits were measured on the first two-root orders based on the root order classification (Pregitzer et al., 2002). Prior to the measurements, root systems were gently washed in deionized water to remove the soils adhering to roots. After the dissection, mixtures of the first two-order roots in each plot/subsamples of each species were arranged and scanned on a LiDE 220 scanner (Canon LiDE 220) at 600 dpi. Thereafter, the subsamples were oven-dried at 65°C for 48 hr and grounded to fine powder after weighing. Root carbon and nitrogen contents were determined using an elemental analyzer (Vario EL III; Elementar). Average root diameter, total root length, surface area and volume were determined using the scanned images by the software of WinRHIZO (Regent Instruments Inc.). SRL was calculated as root total length divided by its dry mass. RTD was obtained as the ratio of root dry mass to its volume. SRA was defined as total surface area divided by its dry mass. Root anatomical traits were determined following the protocols described by Guo et al. (2008). Briefly, the root samples fixed in FAA were gently washed in deionized water prior to the measurement. For the manipulated precipitation change experiment, around 10 root segments of the first two orders of each species in each plot were randomly chosen for anatomical analysis. Along the precipitation gradient transect, about 25 root segments of the first two orders of each species in each site were randomly chosen for anatomical analysis. The segments were embedded in paraffin individually after dehydration by immersion in a sequence of alcohol solution. The roots were then cut into sections with 8 μm thickness. The sections were then stained by the safranine-fast green. The cortex and stele were stained to be green and red, respectively (Kong et al., 2014). The stained roots were photographed using a camera (NIS Elements D 3.0) in conjunction with the microscope (Nikon, 80i). Stele diameter, cortex thickness and vessel diameter were measured using ImageJ (NIH I mage). The number of vessel was counted manually. The ADs of vessels and the ratio of stele diameter to root diameter and to cortex thickness were calculated. The vessel density was calculated as the number of vessels per unit stele cross-sectional area (Long, Kong, Chen, & Zeng, 2013). The measured root traits were list in Table 1. In the manipulated experiment simulating precipitation change, two models were used to detect the effects of water addition on each root traits at species level. Model 1 was a linear model using the “lm” function, where root traits were dependent variables and treatment was independent variable. Model 2 was a linear mixed model using the “lmer” function in “lme4” package, where root traits were dependent variables, treatment was as a fixed factor, and plots were considered as random factors. Then, the “ANOVA” function was used to compare the two models and to determine the better fit. If the two models differ significantly, we determined that the random effects of plots on root traits were significant. The model with lower AIC (Akaike information criterion) was thought to be better fit our data and was chosen to detect the significance of root traits in response to water addition. The statistical analyses were performed using the software package R 3.5.3 (R Core Team, 2015). Then, we used repeated measurement ANOVA and t test with SPSS 19.0 software (SPSS Inc) to estimate the effect of water addition on soil volumetric water content and soil inorganic N (NH4+–N and NO3−–N) concentrations, respectively. Along the precipitation gradients, one-way ANOVA was used to evaluate the effect of sample sites (annual precipitation) on root traits of L. chinensis and A. frigida separately. Thereafter, the least square difference (LSD) tests were used to conduct post hoc mean comparisons of each root traits of the two species in different sites along the precipitation gradients. These analyses were conducted with SPSS 19.0 software (SPSS Inc). Soil volumetric water contents were significantly enhanced by water addition (p < .01) in July and August 2014 (Figure 1a). The increased precipitation had no significant effects on NH4+–N (Figure 1b, p > .05). By contrast, there was a significant increase (3.29 ± 0.90 mg/kg) in soil NO3−–N concentration by water addition (Figure 1b, p < .01). The results of ANOVA showed no significant differences between the two models in the effects of water addition on root traits for most of the case (Table S4), suggesting that the random effects of plots are negligible for most of the case. If there were significant differences between the two models, results from the model showing the lower AIC were used to estimate the significance of the response. The significances of the response of root traits to water addition within each species are shown in Table S5. The root trait values (means ± SE) of the 15 common species in control and water addition plots are shown in Table S6. Root traits of the 15 herbaceous species differed in their directions and magnitudes in response to water addition (Figure 2), and the differences in most root traits between control and water addition treatments were not significant (Figure 2, Tables S5 and S6). Only a few root traits of particular species responded significantly to water addition (Figure 2, Table S5). For example, some grass species showed an increase in density or number of vessels with increasing water supply (Figure 2a,b, p < .05). The proportion of cortex of some leguminous species was increased by water addition (Figure 2k, l, p < .05). Water supply significantly increased RTD and VD in A. frigida, but it reduced SRL and SRA in A. frigida (Figure 2f, p < .05). The SRA of the two liliaceous species showed an opposite trend in response to water addition (Figure 2m,n, p < .05). These results show that different species can adapt to changing precipitation regimes by modulating one or two root traits. The regional common species L. chinensis mainly occurred in sites 1–7 and 10, while A. frigida mainly occurred in sites 3 and 6–10. Root traits of the two species differed among the sampling sites along the precipitation gradients (Table 2), indicating an intraspecific variation in root traits along the precipitation gradients for both species. Root traits values (means ± SE) of the two species along the precipitation gradients are shown in Table S7. The results of LSD indicate that root traits of L. chinensis and A. frigida had different patterns of variation along the precipitation gradients (Figures 3 and S2). For L. chinensis, there was little variation in root AD (Figure 3a), while the other three morphological traits co-varied along the precipitation gradients, with higher SRL and SRA, but lower RTD at sites 6, 7 and 10 (Figure 3b–d). There was no significant difference in RCC along the precipitation gradients except the plants at site 1 (Figure 3e). L. chinensis at sites 5, 6, 7 and 10 exhibited higher root N content (RNC) than those at sites 1, 2, 3 and 4 (Figure 3f). The pattern of variation in root C/N ratio (RCN) along the precipitation gradients was opposite to that of RNC (Figure 3g). The variation in anatomical traits along the precipitation gradients was moderate, as little difference was detected among the anatomical traits at different sites (Figure S2a–g). For A. frigida, the morphological traits co-varied along the precipitation gradients, with higher AD but lower SRL and SRA at sites 9 and 10 (Figure 3h,i,k). Plants of A. frigida at site 10 exhibited higher RTD than plants at other sites (Figure 3j). There was relative small variation in chemical traits of A. frigida (Figure 3l–n). The anatomical traits of A. frigida also showed little variation along the precipitation gradients (Figure S2h–n). In the present study, we discovered that herbaceous species in temperate grasslands can adapt to a range of precipitation change with little root trait plasticity and that there were diverse patterns of variation in root traits of herbaceous species in response to precipitation changes by manipulative experiment at local scale and along the precipitation gradients in temperate steppes of northern China. One important finding is that, despite one or two root traits of several species significant responses to water addition, most root traits of the 15 herbaceous species examined in temperate grasslands are not responsive to water addition (Figure 2, Table S5), highlighting that root traits of the grassland species have low plasticity in response to changing water availability. This finding suggests that root traits have limited plasticity in response to precipitation change in the manipulation experiment at local scale. Theoretically, the phenotypic plasticity of plant traits is mainly determined by plant's internal and ecological factors, including those of the genetic cost, plasticity history and reliability of environmental signal (Nicotra et al., 2010; Valladares et al., 2007). The internal factors usually refer to the phylogenetic constrain on plasticity of root traits (Nicotra et al., 2010; Valladares et al., 2007). Given that most of the root traits are phylogenetically conserved (Chen et al., 2013; Kong et al., 2014; Valverde-Barrantes, Freschet, Roumet, & Blackwood, 2017; Zhou, Bai, et al., 2018), short-term environmental changes may have little impact on root traits immediately. This explanation is also consistent with that of Zadworny et al. (2016) and Wang et al. (2017) reporting that root traits of Scots pine and plant species in subtropical and boreal forests were mainly restricted by their phylogeny, and that root traits had limited plasticity in response to changing environments. The low plasticity in root traits of the herbaceous species may be an adaptive trait to the local environments. In the temperate steppes of our studies, annual precipitation displayed substantial inter-annual fluctuation (Figure S3), ranging from 196.3 to 384.8 mm between 2006 and 2015. The large inter-annual fluctuations of precipitations may lead to wide threshold intervals of the root trait values to ensure an efficient acquisition of resource under variable water availabilities, thus allowing them to buffer against the inter-annual variation of precipitation. In our studies, the amount of water added was 30% of the local annual precipitation, which is similar to the naturally occurred inter-annual fluctuation of precipitation. Thus, the plant species may exhibit lower phenotypic plasticity to water addition. Moreover, the small magnitude of response to changing precipitation can enable the stable plant development under the variable environmental conditions (Freschet, Cornelissen, Logtestijn, & Aerts, 2010). At regional scale, root traits of both L. chinensis and A. frigida varied along the precipitation gradients, indicating a plastic adjustment to the changing precipitation (Table 2, Figure 3). However, the results of LSD showed that the phenotypic root traits of both species had no significant differences at higher precipitation regions. In contrast with our first hypothesis, these results indicate that the responses of root traits to long-term precipitation change (along the precipitation gradients) were similar to the short-term precipitation change (water addition experiments), and that certain root traits of a specific species can allow for adaptation of plants to a range of precipitation changes at both regional and local scales. Moreover, the differences in root morphological and chemical traits only reached to the significant level when the precipitation was lower than a certain value, that is 250 and 160 mm for L. chinensis and A. frigida, respectively (Figure 3). These results suggest that the phenotypic adjustment of adaptation may occur only when the range of annual precipitation exceeds a certain threshold. The little plasticity of root traits under a certain range of annual precipitation both in manipulation experiment and along the precipitation gradients may indicate that plants in semi-arid grasslands have the capacity to buffer a range of annual precipitation change. This capacity may ensure the resource acquisition under variable soil resources, thus conferring the stable plant development of herbaceous species to withstand the large inter-annual fluctuation of precipitation in semi-arid grasslands. Moreover, the characteristics of herbaceous species in grasslands may also allow a sustainable ecosystem functions in future climate change. The variation patterns in root traits differed significantly between L. chinensis and A. frigida both in the manipulation experiment and along the precipitation gradients, highlighting that alternative strategies may have been evolved by different species to adapt to the changing precipitation. In the manipulation experiment, the root traits of L. chinensis were not significantly responsive to water addition, while A. frigida adjusted the morphological traits in response to water addition (Figure 2d,f). We also demonstrated that, along the precipitation gradients, L. chinensis had higher SRL, SRA and RNC, as well as higher proportion of stele with more and larger vessel at sites with the annual precipitation lower than c. 250 mm, e.g. at sites 6, 7 and 10 (Figure 3). While site 5 had a similar MAP with sites 6 and 7, the root traits of L. chinensis at site 5 showed a different pattern. This may result from the lower altitude and higher MAT at site 5 compared with sites 6 and 7. In contrast, A. frigida had higher root diameter and RTD, but lower SRL and SRA at sites with the annual precipitation lower than c. 160 mm (e.g. at sites 9 and 10) (Figure 3). These differences in phenotypic adaptation of root traits may reflect their different adaptive strategies between the two species to the episodic nature of water input in the arid and semi-arid ecosystems. In arid, water-limited environments, water supply often displays pulsed patterns. For example, water input due to erratic rainfall events was always followed by long droughts, particularly in the drier environments (Austin et al., 2004; Schwinning, Sala, Loik, & Ehleringer, 2004). Under these conditions, plant species have to deal with the trade-off between water exploitation during the wet period and drought tolerance during the dry span (Méndez-Alonzo, Paz, Cruz-Zuluaga, Rosell, & Olson, 2012; Pineda-Garcia et al., 2016; Pineda-GarcÍA, Paz, & Tinoco-Ojanguren, 2011). The phenotypic plasticity of L. chinensis seems to be water exploitation (acquisitive) strategy under drier environments, such that the higher SRL and larger vessel can enhance water and nutrient uptake and transport, thus leading to fast growth under the conditions of pulsed precipitation (Eissenstat, 1991; Kong et al., 2016; Padilla et al., 2013). In contrast, thinner roots combined with lower RTD would decrease the defenses and fitness of plants, resulting in faster turnover rates to cope with prolonged and/or severe drought (Comas & Eissenstat, 2004; Eissenstat, Wells, Yanai, & Whitbeck, 2000; Larson & Funk, 2016). Thus, rapid acquisition to exploit transient periods of high soil water availability and fast turnover may be an important strategy for survival of L. chinensis in the drier environments of semi-arid grasslands. On the other hand, the phenotypic plasticity of A. frigida is likely to be drought tolerance (conservative) strategy to adapt to drier and lower precipitation conditions. This is because the higher diameter and tissue density can promote penetrative ability as well as the drought tolerance in dry soils (Pineda-Garcia et al., 2016). In contrast, the lower SRL and area can minimize the loss of plant water and nutrients (Clark, Whalley, & Barraclough, 2003; Hernandez, Vilagrosa, Pausas, & Bellot, 2010; Wright & Westoby, 1999). These strategies may also result from the adaptions to other biotic and/or abiotic factors, such as soil C:N (Ostonen et al., 2017). However, in the present study, we mainly focused on the MAP as the gradients of MAP are continuous and dominant compared with other abiotic factors along the precipitation gradients. Similar to the diverse variation patterns along the precipitation gradients, an increase in water availability led to changes in several root traits of plant species in the manipulation experiments at local scale (Figure 2). These results support our second hypothesis that the response of root traits to water availability was species-specific and that trait-dependent adaptation to water availability existed in both manipulated precipitation change experiment at local scale and along the precipitation gradients. In this context, Fry, Evans, Sturrock, Bullock, and Bardgett (2018) reported the different plasticity of root in response to draught among plant species with different root architectures. These findings indicate that different plant species have evolved diverse adaptive strategies to cope with the changing regimes of water availability. These results are in agreement with previous studies reporting multifarious trait-based adaptive strategies among species (Larson & Funk, 2016; Li, Liu, McCormack, Ma, & Guo, 2017; Tobner, Paquette, & Messier, 2013). The species-specific and trait-dependent response of plant roots to changing precipitation may enable different species occupy different trait space and niches, thus ensuring full utilization of resources and coexistence under changing climate conditions (Freschet, Bellingham, Lyver, Bonner, & Wardle, 2013; Laughlin & Messier, 2015; Li et al., 2017). Our results for the response of root traits at species level may provide important information on plants’ response in the community under future climate change. Our results reveal that most of the response of root traits to water addition at species level was not significant among the 15 herbaceous species in temperate steppes. The root traits of the two regionally common species along the precipitation gradients varied distinctly at sites with annual precipitation lower than a certain value. These results indicate that the plasticity of root traits is limited, and plants can adapt to a range of precipitation change, and that the adaption may occur only when the range of annual precipitation exceeds certain thresholds. In addition, the patterns of variation in root traits of the two regionally common species (L. chinensis and A. frigida) differed both in manipulation experiments and along the precipitation gradients, and significant responses to water addition of one or two root traits of several species were detected. These results indicate that root traits in response to precipitation change are species-specific and trait-dependent, and that diverse adaptive strategies exist among different species. Our findings also highlight the complexity of root trait response to water availability even when the methods are standardized. The studies on the response of root traits at species level may provide important information on predicting plants response to future climate change. We thank Guangquan Wang and Yunfeng Dong for their help in field and laboratory work. This research was supported by the National Natural Science Foundation of China (31670481, 31370468) and the State Key Basic Research Development Program of China (2016YFC0500706). W.B. conceived the ideas and designed methodology; M.Z., J.W. and Y.Z. collected the data; M.Z. and J.W. analysed the data; M.Z., W.B. and W.-H.Z. led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication. All data presented in this manuscript are available from the Dryad Digital Repository https://doi.org/10.5061/dryad.45m46kp (Zhou, Wang, Bai, Zhang, & Zhang, 2019). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article." @default.
• W2964646550 created "2019-08-13" @default.
• W2964646550 creator A5037677450 @default.
• W2964646550 creator A5042084246 @default.
• W2964646550 creator A5044320922 @default.
• W2964646550 creator A5058488226 @default.
• W2964646550 creator A5062647006 @default.
• W2964646550 date "2019-08-22" @default.
• W2964646550 modified "2023-10-05" @default.
• W2964646550 title "The response of root traits to precipitation change of herbaceous species in temperate steppes" @default.
• W2964646550 cites W1578144663 @default.
• W2964646550 cites W1600427857 @default.
• W2964646550 cites W1959243054 @default.
• W2964646550 cites W1966895245 @default.
• W2964646550 cites W1971852746 @default.
• W2964646550 cites W1976988124 @default.
• W2964646550 cites W1990236771 @default.
• W2964646550 cites W1990460478 @default.
• W2964646550 cites W1991240373 @default.
• W2964646550 cites W1998307303 @default.
• W2964646550 cites W2002260830 @default.
• W2964646550 cites W2006694842 @default.
• W2964646550 cites W2018757639 @default.
• W2964646550 cites W2024926484 @default.
• W2964646550 cites W2040087381 @default.
• W2964646550 cites W2061070657 @default.
• W2964646550 cites W2066050444 @default.
• W2964646550 cites W2069760915 @default.
• W2964646550 cites W2070138228 @default.
• W2964646550 cites W2078405076 @default.
• W2964646550 cites W2083521072 @default.
• W2964646550 cites W2096212927 @default.
• W2964646550 cites W2096567926 @default.
• W2964646550 cites W2101364071 @default.
• W2964646550 cites W2102633940 @default.
• W2964646550 cites W2104804673 @default.
• W2964646550 cites W2116927862 @default.
• W2964646550 cites W2117631271 @default.
• W2964646550 cites W2119510269 @default.
• W2964646550 cites W2126559074 @default.
• W2964646550 cites W2127928904 @default.
• W2964646550 cites W2128440160 @default.
• W2964646550 cites W2131805332 @default.
• W2964646550 cites W2132662500 @default.
• W2964646550 cites W2134143138 @default.
• W2964646550 cites W2134800114 @default.
• W2964646550 cites W2140277878 @default.
• W2964646550 cites W2151303612 @default.
• W2964646550 cites W2152895243 @default.
• W2964646550 cites W2154524337 @default.
• W2964646550 cites W2162199816 @default.
• W2964646550 cites W2163143186 @default.
• W2964646550 cites W2163688790 @default.
• W2964646550 cites W2169651581 @default.
• W2964646550 cites W2170193083 @default.
• W2964646550 cites W2171753234 @default.
• W2964646550 cites W2172167884 @default.
• W2964646550 cites W2249700495 @default.
• W2964646550 cites W2322837830 @default.
• W2964646550 cites W2388386570 @default.
• W2964646550 cites W2442638381 @default.
• W2964646550 cites W2546013994 @default.
• W2964646550 cites W2550975580 @default.
• W2964646550 cites W2561848104 @default.
• W2964646550 cites W2580026137 @default.
• W2964646550 cites W2592423967 @default.
• W2964646550 cites W2604440501 @default.
• W2964646550 cites W2608589959 @default.
• W2964646550 cites W2622044568 @default.
• W2964646550 cites W2734486312 @default.
• W2964646550 cites W2741864438 @default.
• W2964646550 cites W2756029367 @default.
• W2964646550 cites W2765305249 @default.
• W2964646550 cites W2789887632 @default.
• W2964646550 cites W2791775578 @default.
• W2964646550 cites W2796474941 @default.
• W2964646550 cites W2800572433 @default.
• W2964646550 cites W2806079555 @default.
• W2964646550 cites W2806766638 @default.
• W2964646550 cites W2886802822 @default.
• W2964646550 cites W2890406182 @default.
• W2964646550 cites W3036727350 @default.
• W2964646550 cites W4239471541 @default.
• W2964646550 doi "https://doi.org/10.1111/1365-2435.13420" @default.
• W2964646550 hasPublicationYear "2019" @default.
• W2964646550 type Work @default.
• W2964646550 sameAs 2964646550 @default.
• W2964646550 citedByCount "29" @default.
• W2964646550 countsByYear W29646465502020 @default.
• W2964646550 countsByYear W29646465502021 @default.
• W2964646550 countsByYear W29646465502022 @default.
• W2964646550 countsByYear W29646465502023 @default.
• W2964646550 crossrefType "journal-article" @default.
• W2964646550 hasAuthorship W2964646550A5037677450 @default.
• W2964646550 hasAuthorship W2964646550A5042084246 @default.
• W2964646550 hasAuthorship W2964646550A5044320922 @default.
• W2964646550 hasAuthorship W2964646550A5058488226 @default.
• W2964646550 hasAuthorship W2964646550A5062647006 @default.

• next