Effects of Warming and Increased Precipitation on Root Production and Turnover of Stipa breviflora Community in Desert Steppe

: Organic carbon in grassland mainly exists in the soil, and root production and turnover play important roles in carbon input into the soil. However, the effects of climate change on plant root dynamics in desert steppe are unknown. We conducted an experiment in a desert steppe, which included ambient temperature (T0); temperature increased by 2 ◦ C (T1); temperature increased by 4 ◦ C (T2); natural precipitation (P0); precipitation increased by 25% (P1); precipitation increased by 50% (P2); and the interaction between warming and increased precipitation. Plant community aboveground characteristics; root production; and root turnover were measured. We found that the root length production of the T0P2; T1P1; T2P0; and T2P1 treatments were significantly higher than that of the T0P0 treatment, with an increment of 98.70%, 11.72%, 163.03%, and 85.14%, respectively. Three treatments with temperature increased by 2 ◦ C (T1P0; T1P1; and T1P2) and significantly increased root turnover rate compared to the T0P0 treatment, with increases of 62.53%, 42.57%, and 35.55%, respectively. The interaction between warming and increased precipitation significantly affected the root production of the community ( p < 0.01), but this interaction was non-additive. Future climate warming will benefit the accumulation of root-derived carbon in desert steppe communities.


Introduction
Due to human activities emitting large amounts of greenhouse gases since the Industrial Revolution, global and regional climate change has been exacerbated, mainly climate warming and extreme precipitation.It is estimated that the global surface temperature will rise by 1.0-3.7 • C by the end of the 21st century [1].Global warming is accelerating hydrological cycles, leading to significant changes in precipitation patterns in terrestrial ecosystems, particularly in high-latitude regions [2].China has been continuously warming with a temperature of 0.39 • C/decade, with significant warming occurring especially in northwestern China, including Inner Mongolia [3].Annual precipitation in the northwest region of China has significantly increased at a rate of 0.55 mm/year [4].Temperature and moisture determine most of the processes involved in photosynthetic carbon sequestration, including photosynthesis, carbon distribution in the aboveground and belowground parts, transpiration, and root growth during plant growth [5].
The Inner Mongolia desert steppe is an indispensable part of the temperate steppe of Central Asia [6,7].Desert steppes belong to the transitional zone between desert and typical steppes.As a result of harsh ecological environments and climate fluctuation, the desert steppe ecosystem is very fragile, which not only has an impact on the production and life of Inner Mongolia but also has a crucial influence on the carbon cycle in China and even the world [8].Fine roots play an extremely important role in the carbon exchange process that connects the aboveground parts and the soil [9].The root turnover rate is often defined as the number of times per year that fine root biomass is replaced [10].The rate of root turnover is closely related to biochemical cycling in terrestrial ecosystems [11].Plants can produce and store carbohydrates synthesized by photosynthesis through their roots and then transfer carbon from dead roots to the soil through root turnover [12].The net primary productivity (NPP) of ecosystems includes aboveground net primary productivity (ANPP) and belowground net primary productivity (BNPP) [13].The distribution of NPP between aboveground and belowground components is an important process in the carbon cycle and energy flow of ecosystems and is also one of the main components of the global carbon budget [14,15].Climate change directly or indirectly affects NPP, so understanding NPP dynamics is essential to enhance our understanding of the impact of grassland belowground carbon allocation and carbon storage.Because of this, a great number of climate change experiments have been carried out, such as studies showing how plant root production will increase [16], decrease [17], or remain unchanged [18] due to climate warming.Changes in precipitation also strongly influence changes in root production, with reduced precipitation increasing root production in alpine steppes and increased precipitation increasing root production in typical steppes.Differences between the results of different studies may be due to their different regional environments due to factors such as climatic conditions, vegetation types, and altitude.
Climatic conditions are considered to be the main drivers of root growth and turnover [19], and a meta-analysis of the world's grasslands found that temperature and precipitation had a stronger impact on root turnover than soil physico-chemical properties [20].Climate warming increases soil net mineralization rate and reduces soil water availability, which may affect root productivity and turnover [21,22].Temperature and precipitation are necessary for plant growth and development, but the effects are still unclear.Higher temperatures lead to lower soil water content [23], which exacerbates drought stress, so plants allocate more carbon to root production to obtain an equal amount of water, thus promoting root growth [24].In the semi-arid temperate steppe of northern China, it was also concluded that warming promoted root production and turnover [25].Therefore, our first hypothesis is that warming will increase root production and turnover of Stipa breviflora communities in the desert steppe examined in this study.Increased precipitation will increase soil moisture content, so our second hypothesis is that increased precipitation within a certain range will reduce the root production and turnover of Stipa breviflora communities in the desert steppe.Climate change experiments on the temperate steppe in northern China have shown that warming and increased precipitation have a significant interactive effect on root production and mortality [25].Thus, our third hypothesis is that there is an interaction between warming and increased precipitation on root production and turnover in the desert steppe.The objective of this study is to characterize the root production and turnover of plant communities in the desert steppe and to evaluate the effects of warming and increased precipitation on plant community aboveground characteristics, root production, and root turnover.

Study Site
The field experiment site is located in Siziwang Banner (41 • 47 ′ 19 ′′ N, 111 • 53 ′ 45 ′′ E, 1456 m elevation).This region has a typical temperate continental monsoon climate, with cold and long winters, hot summers, and a great temperature difference between day and night.The mean annual precipitation is 280 mm, with precipitation mostly occurring from May to September.The mean annual temperature is 7.2 • C. The soil in the region is light castanozems with a lack of phosphorus, as well as less nitrogen, more potassium, and low organic matter content (the total phosphorus content was 0.31 g/kg, the total nitrogen content was 1.27 g/kg, the total potassium content was 44.07 mg/kg, and the soil organic matter content was 19.27 g/kg) [26].The vegetation type in the region is that of the desert steppe, and the constructive species is Stipa breviflora; dominant species include Cleistogenes songorica and Artemisia frigida, and associated species include Convolvulus ammannii and Kochia prostrata and other plants.

Experimental Design
The experiment commenced in 2014, and an open-top chamber (OTC) was used to increase temperature.The OTC used was a hexagonal open-top chamber with a side length of 1.5 m and two kinds of height, 1 m and 2.3 m (1 m and 2.3 m can increase temperature by 1-2 • C and 2-4 • C, respectively), with the top design being a fully or partial opening.The OTC was set up with a 3 m diameter, and a 1 m × 1 m quadrat placed in the center [27].Precipitation increase was achieved by using rainwater interceptors with 25% and 50% of the basal area of the OTC, respectively.After each rainfall, the rainwater collected in the interceptors was poured into the corresponding plots by the workers (Figure 1a).The experiment treatments included three temperature gradients, ambient temperature (T0), temperature increased by 1-2 • C (T1), and temperature increased by 2-4 • C (T2), as well as three precipitation gradients, natural precipitation (P0), precipitation increased by 25% (P1), and precipitation increased by 50% (P2), with a total of nine treatments (i.e., T0P0, T0P1, T0P2, T1P0, T1P1, T1P2, T2P0, T2P1, and T2P2) and four replicates for each treatment, with a total of 36 plots (Figure 1b).
nitrogen content was 1.27 g/kg, the total potassium content was 44.07 mg/kg, and the organic matter content was 19.27 g/kg) [26].The vegetation type in the region is th the desert steppe, and the constructive species is Stipa breviflora; dominant species inc Cleistogenes songorica and Artemisia frigida, and associated species include Convolv ammannii and Kochia prostrata and other plants.

Experimental Design
The experiment commenced in 2014, and an open-top chamber (OTC) was use increase temperature.The OTC used was a hexagonal open-top chamber with a length of 1.5 m and two kinds of height, 1 m and 2.3 m (1 m and 2.3 m can incr temperature by 1-2 °C and 2-4 °C, respectively), with the top design being a ful partial opening.The OTC was set up with a 3 m diameter, and a 1 m × 1 m quadrat pl in the center [27].Precipitation increase was achieved by using rainwater interceptors 25% and 50% of the basal area of the OTC, respectively.After each rainfall, the rainw collected in the interceptors was poured into the corresponding plots by the wor (Figure 1a).The experiment treatments included three temperature gradients, am temperature (T0), temperature increased by 1-2 °C (T1), and temperature increased by °C (T2), as well as three precipitation gradients, natural precipitation (P0), precipit increased by 25% (P1), and precipitation increased by 50% (P2), with a total of treatments (i.e., T0P0, T0P1, T0P2, T1P0, T1P1, T1P2, T2P0, T2P1, and T2P2) and replicates for each treatment, with a total of 36 plots (Figure 1b).
Stipa breviflora was selected as the research object in the quadrat, and a 20 cm soil hole was drilled vertically next to it.An organic glass tube with a diameter of 7. was buried.The tube was marked at 10 cm and 20 cm below the ground with a b line.The exposed part of the tube was painted black and covered with an opaque lid  Stipa breviflora was selected as the research object in the quadrat, and a 20 cm deep soil hole was drilled vertically next to it.An organic glass tube with a diameter of 7.5 cm was buried.The tube was marked at 10 cm and 20 cm below the ground with a black line.The exposed part of the tube was painted black and covered with an opaque lid.

Data Collection and Calculation
Image collection was carried out from May to October 2019 using a root scanner (CI-600, CID Bio-Science Inc., Camas, WA, USA), scanning every 15 days with a resolution of 11.6 pixels per mm.Rooyfly software (Version 2.0.2,Rootfly Development Team, Clemson, SC, USA) was used to process the scanned map for root growth.A total of 360 images were collected during the growth season.In total, 3673 roots were tracked and marked.Dead and living roots were identified by the color of the root, all black and disappearing roots were recorded as dead, and light brown and white roots were recorded as alive.
The calculations of root production, root standing crop, and root turnover mainly follow Bai and Huo et al. [25,28].Root production was calculated by the sum of the length of all new roots and the extension length of previously existing roots in each sampling period.Annual mean root standing crop were calculated by the average length of live roots at each observing date of a year.Root turnover rate was calculated by annual root length production, dividing it by annual mean root standing crop Plant community surveys were conducted in August 2019, with one permanent 1 m × 1 m quadrat in each experimental replicate, where the height, cover, density, and frequency of all species within the quadrats were recorded.Since the experimental restrictions did not allow for the destruction of the plants within the quadrats, 5-10 plants of equal size were selected around the quadrats to be mowed with the ground based on the results of the recordings and were brought back to the laboratory to be dried and weighed before calculating the aboveground production.We used Equation ( 1) to calculate the importance values of plant species in the quadrats [29].
Hr, Cr, Dr, Fr, and Br are the relative height, cover, density, frequency, and biomass, respectively.
Meteorological data were collected by a meteorological station (Dynamet, Dynamax Inc., CA, USA) in the experiment site.
Soil temperature and humidity were measured in the topsoil (0-10 cm), using a soil temperature and humidity meter (HH2 Moisture Meter, Delta-T Devices Ltd., Cambridge, UK), at noon on the same day as root observation, once in each plot.

Data Analysis
The Shapiro-Wilk test and Levene test were used to test the normality hypothesis and the homoscedasticity hypothesis of the residuals.The differences in soil temperature, soil moisture, ANPP, root production, root standing crop, and root turnover rate among the treatments were tested using both LSD and post hoc Duncan tests for multiple comparisons, with a significance level of 0.05.Two-way ANOVA was employed to compare the effects of warming, increased precipitation, and their interaction on the ANPP, root production, and root turnover rate (Duncan, p < 0.05).A structural equation model was executed in AMOS 21.0.First, possible explanatory variables were included in the path analysis by reviewing the relevant literature on plant root production and turnover.Secondly, we used a maximum likelihood estimation technique to obtain path coefficients and then eliminated the insignificant paths until we obtained the final model.Finally, we used a Chi-square test and the root mean square error of approximation to evaluate the fit of the model.All statistical analyses were conducted with IBM SPSS Statistics 26.Figures were elaborated using Origin 2021, and tables were elaborated using Microsoft Excel 2019.

Seasonal Variations in the Root Length Production and Root Standing Crop of Plant Communities
Root length production was greatest in August during the experimental period (Figure 2).The root length production of the four treatments with T2P0, T2P1, T2P2, and T1P2 showed a trend of decreasing first in July and then increasing with a peak in production in August.
T1P2 showed a trend of decreasing first in July and then increasing with a peak in production in August.
The peaks of the root standing crop appeared from the 22nd of July to the 8th of August, and the peaks of the T1P0 treatment, the T2P0 treatment, and the T2P2 treatment were in early July; the peak of the T0P2 treatment was in early August.The root standing crop of all the treatments showed a trend of first increasing and then decreasing throughout the experimental period, and the three treatments with precipitation increased by 50% (T0P2, T1P2, and T2P2), showing an increasing trend in the second half of August (Figure 3).The peaks of the root standing crop appeared from the 22nd of July to the 8th of August, and the peaks of the T1P0 treatment, the T2P0 treatment, and the T2P2 treatment were in early July; the peak of the T0P2 treatment was in early August.The root standing crop of all the treatments showed a trend of first increasing and then decreasing throughout the experimental period, and the three treatments with precipitation increased by 50% (T0P2, T1P2, and T2P2), showing an increasing trend in the second half of August (Figure 3).T1P2 showed a trend of decreasing first in July and then increasing with a peak in production in August.
The peaks of the root standing crop appeared from the 22nd of July to the 8th of August, and the peaks of the T1P0 treatment, the T2P0 treatment, and the T2P2 treatment were in early July; the peak of the T0P2 treatment was in early August.The root standing crop of all the treatments showed a trend of first increasing and then decreasing throughout the experimental period, and the three treatments with precipitation increased by 50% (T0P2, T1P2, and T2P2), showing an increasing trend in the second half of August (Figure 3).

Effects of Different Treatments on ANPP and Root Length Production
Warming, increased precipitation, and their interaction significantly affected ANPP (p < 0.001, Table 1).ANPP in all treatments was 132.34 ± 6.57 g/m 2 (Figure 4).The T1P0 and T2P2 treatment significantly increased community ANPP (p < 0.05) in comparison to the T0P0 treatment, with an increase of 49.20% and 79.14%, respectively.increase of 98.70%.The root length production of T1P1 was significantly higher tha of the T1P2 treatment and the T0P0 treatment (p < 0.05), with an increment of 11 and 111.72%, respectively.Under the condition of warming by 4 °C, with the incre precipitation, root length production showed a decreasing trend, and the root l production of T2P2 treatment was significantly lower than that in the T2P0 treatmen 0.05), with a decrease of 78.45%.The T2P0 treatment and T2P1 treatment signific increased root length production (p < 0.05) in comparison to the T0P0 treatment, w increase of 163.03% and 85.14%, respectively.The interaction of warming and increased precipitation significantly affected the root length production of the community in the growing season (p < 0.01, Table 1).In the 0-20 cm soil, the root length production of all the treatments was 45.14 ± 3.11 m/m 2 (Figure 4).Under natural temperature conditions, with the increase in precipitation, root length production showed an increasing trend, and root length production of the T0P2 treatment was significantly higher than that of the T0P0 treatment (p < 0.05), with an increase of 98.70%.The root length production of T1P1 was significantly higher than that of the T1P2 treatment and the T0P0 treatment (p < 0.05), with an increment of 116.15% and 111.72%, respectively.Under the condition of warming by 4 • C, with the increase in precipitation, root length production showed a decreasing trend, and the root length production of T2P2 treatment was significantly lower than that in the T2P0 treatment (p < 0.05), with a decrease of 78.45%.The T2P0 treatment and T2P1 treatment significantly increased root length production (p < 0.05) in comparison to the T0P0 treatment, with an increase of 163.03% and 85.14%, respectively.

Effects of Different Treatments on Root Turnover Rate of Plant Community
Root turnover rate tended to decrease with increasing precipitation under the same warming conditions (Figure 5).Both warming and increased precipitation significantly affected root turnover (p < 0.01, Table 1).In the 0-20 cm soil, the root turnover rate of all the treatments was 1.98 ± 0.11 year −1 .Under the condition of warming by 2 • C, the three precipitation treatments (T1P0, T1P1, and T1P2) significantly increased the root turnover rate (p < 0.05) in comparison to the T0P0 treatment, with increases of 62.53%, 42.57%, and 35.55%, respectively.Under natural precipitation conditions, warming resulted in an increasing trend in root turnover, which diminished with increasing temperatures.Root turnover showed a decreasing trend with increasing precipitation under warming conditions, and this decreasing trend was more pronounced at higher temperatures.
Root turnover rate tended to decrease with increasing precipitation under the warming conditions (Figure 5).Both warming and increased precipitation signific affected root turnover (p < 0.01, Table 1).In the 0-20 cm soil, the root turnover rate of a treatments was 1.98 ± 0.11 year −1 .Under the condition of warming by 2 °C, the precipitation treatments (T1P0, T1P1, and T1P2) significantly increased the root turn rate (p < 0.05) in comparison to the T0P0 treatment, with increases of 62.53%, 42.57% 35.55%, respectively.Under natural precipitation conditions, warming resulted i increasing trend in root turnover, which diminished with increasing temperatures.turnover showed a decreasing trend with increasing precipitation under war conditions, and this decreasing trend was more pronounced at higher temperatures.
Through the structural equation model, it can be seen that the correlation bet root turnover and environmental factors (temperature and precipitation) was (Figure 6).Warming had a significant positive effect on the root turnover rate of the breviflora community (effect value of 0.332); however, increased precipitation h significant negative effect on the root turnover of the Stipa breviflora community important value of Stipa breviflora (effect values of −0.678 and −0.404).The effe precipitation increase on the root turnover rate of the Stipa breviflora community greater than that of warming.Through the structural equation model, it can be seen that the correlation between root turnover and environmental factors (temperature and precipitation) was high (Figure 6).Warming had a significant positive effect on the root turnover rate of the Stipa breviflora community (effect value of 0.332); however, increased precipitation had a significant negative effect on the root turnover of the Stipa breviflora community and important value of Stipa breviflora (effect values of −0.678 and −0.404).The effect of precipitation increase on the root turnover rate of the Stipa breviflora community was greater than that of warming.

Effects of Warming and Increased Precipitation on the Root Length Production of the Stip breviflora Community
The results of this study only partially support our first and second hypotheses.positive effects of warming on root length production and turnover rate were obser under natural precipitation treatments, which is consistent with other results obtained warming experiments [19,25].However, climate warming has an inhibitory effect these variables under the conditions of increased precipitation.We found a signific interaction between warming and increased precipitation on root production a standing crop, which is partly consistent with the third hypothesis.The structu equation model showed that warming had a negative correlation with root producti and increased precipitation had a positive correlation with root production.The stro interaction between warming and increased precipitation implied that warming a increased precipitation in the desert steppe had a non-additive relationship with r production in the Stipa breviflora community.Because the increased temperatures accelerate the rate of soil moisture evaporation [23], plants grow in order to absorb m water, which in turn promotes the growth of the primary root to deeper soil [9,24] environments with relatively abundant precipitation, the positive impact of clim change may be stronger than the negative impact, thereby promoting root product [30].However, in arid or semi-arid ecosystems, moisture is the primary limiting fac which limits the root production in desert steppe ecosystems, and the negative impac climate warming may be stronger than the positive impact [25].This study indicated t warming by 1.5-2 °C had a negative effect on root production and standing crop in alp meadows [18], and similar conclusions were also drawn in typical temperate step [25].In addition, temperature and precipitation had a significant interaction and effect root production, but precipitation had different effects on root production un different temperature conditions.Because different magnitudes of warming h different effects on soil moisture content, different magnitudes of precipitation incre  The results of this study only partially support our first and second hypotheses.The positive effects of warming on root length production and turnover rate were observed under natural precipitation treatments, which is consistent with other results obtained by warming experiments [19,25].However, climate warming has an inhibitory effect on these variables under the conditions of increased precipitation.We found a significant interaction between warming and increased precipitation on root production and standing crop, which is partly consistent with the third hypothesis.The structural equation model showed that warming had a negative correlation with root production, and increased precipitation had a positive correlation with root production.The strong interaction between warming and increased precipitation implied that warming and increased precipitation in the desert steppe had a non-additive relationship with root production in the Stipa breviflora community.Because the increased temperatures can accelerate the rate of soil moisture evaporation [23], plants grow in order to absorb more water, which in turn promotes the growth of the primary root to deeper soil [9,24].In environments with relatively abundant precipitation, the positive impact of climate change may be stronger than the negative impact, thereby promoting root production [30].However, in arid or semi-arid ecosystems, moisture is the primary limiting factor, which limits the root production in desert steppe ecosystems, and the negative impact of climate warming may be stronger than the positive impact [25].This study indicated that warming by 1.5-2 • C had a negative effect on root production and standing crop in alpine meadows [18], and similar conclusions were also drawn in typical temperate steppes [25].In addition, temperature and precipitation had a significant interaction and effect on root production, but precipitation had different effects on root production under different temperature conditions.Because different magnitudes of warming have different effects on soil moisture content, different magnitudes of precipitation increase can improve soil moisture limitations [25].This study found that warming, increased precipitation, or their interactions could promote root production, but the effect on the root standing crop was not significant.Because plants constantly adjust their nutrient distribution patterns according to different temperatures and precipitation environments [31], such as increased precipitation in desert steppes, more species will compete for this limited resource.Due to the increase in species, the height, coverage, and density of the community will change [32].Plants with a better environment need to compete for more resources, such as sunlight, CO 2 , moisture, and so on.Therefore, in order to better survive, plants will have different aboveground and belowground growth patterns and nutrient distribution patterns at different times.The difference in root production is the performance of plant roots in adapting to different environments, while the lack of difference in standing crop may be due to the long-term adaptation of plants to desert steppes or the fact that the difference has not yet been shown due to the short treatment time.Because this study only observed the roots in 0-20 cm soil, it may also affect the results of root production and standing crop.However, a study of temperate steppes in northern China (including meadow steppes, typical steppes, and desert steppes) showed that most of the root lengths of the fibrous-rooted plant species were within 0-20 cm, and only a few taproot plant species had a root length longer than 20 cm [33], but these did not belong to the dominant species of desert steppes.Therefore, the root observation in 0-20 cm soil in this study conforms to the majority of root growth ranges.
This study found that increasing precipitation by 50% under warming conditions did not affect root production but increased ANPP.Because substantial warming leads to the evaporation of soil moisture, increased precipitation reduces soil moisture dissipation, lowers soil temperatures, increases plant cover, and promotes aboveground plant growth [34].Small increases in temperature have been shown to increase plant ANPP in American short-grass prairie [35], which is the same as the results of this study.Perhaps it is because slight warming enhances the aboveground productivity of plants by promoting photosynthesis and prolonging their phenology [36].Significant warming promotes the early re-greening of plants, and the appropriate temperature in the early growing season can promote the accumulation of nutrients in plants [36].However, when the temperature is high in the late growing season, significant warming increases the severity of soil drought and slows down plant growth.

Effects of Warming and Increased Precipitation on the Root Turnover of the Stipa breviflora Community
In this study, the root turnover rate was positively correlated with warming and negatively correlated with precipitation increase, which was consistent with the first and second hypotheses.This study also found that warming by 2 • C promoted root turnover.Warming accelerated nutrient mineralization rate and root respiration rate, resulting in shortened optimal root lifespan [37,38], Decayed roots promote the decomposition of roots by microorganisms and return nutrients from the roots to the soil for plant reabsorption [39,40], thereby accelerating nutrient cycling.One study found that the fine root turnover of steppes increases exponentially with global warming [36], which is similar to some of the results of this study.Studies have shown that shrubs have short-lived roots under drought conditions [41], and studies have shown that desert plants grow fine roots quickly when it rains and fall off when it is dry [42], which may be a strategy for plants to cope with drought stress.The changes in the root production and turnover of plant communities in the desert steppe in the future depend on the precipitation variation in the region.From 1960 to 2013, the annual average temperature and rainfall in most regions of northwest China showed an increasing trend [4].According to this trend, it is predicted that the root productivity of communities in the desert steppe in northwest China will increase, but this is also regulated by precipitation; a small increase in temperature and precipitation (a 2 • C and 25% increase in precipitation in this study) is beneficial to the turnover of community roots, but a greater increase in temperature and precipitation (a 4 • C and 50% increase in precipitation in this study) does not affect the turnover of community roots.The experimental site in this study is located in the central part of Inner Mongolia.Based on the meteorological data of Inner Mongolia from 1961 to 2012, the annual rainfall in the central part of Inner Mongolia has not changed after 1975 (except for several wet years), while the temperature has increased by 0.37 • C/decade, which is significantly higher than the global warming rate (0.14 • C/decade) [43].

Conclusions
In the desert steppe of northern China, the main effect of experimental warming on root turnover was positive, and the main effect of increased precipitation on root turnover was negative.This study found that warming and increased precipitation had a significant interaction and effect on root production and standing crop, and the interaction was nonadditive.Based on this trend, it is predicted that future climate change will be beneficial to the root production and turnover of desert steppe communities in Inner Mongolia.These results will provide theoretical assistance for predicting the response of temperate desert steppes to global climate change.

Figure 6 .
Figure 6.Structural equation model of warming and increased precipitation with Stipa breviflora aboveground and belowground characteristics of community.The red and blue lines indi significant positive and negative impacts, respectively.Numbers are standardized path coefficie Goodness-of-fit index: χ 2 is Chi-square, df is degrees of freedom, and RMSEA is the root m square error of approximation.*, **, and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.

Figure 6 .
Figure 6.Structural equation model of warming and increased precipitation with Stipa breviflora and aboveground and belowground characteristics of community.The red and blue lines indicate significant positive and negative impacts, respectively.Numbers are standardized path coefficients.Goodness-of-fit index: χ 2 is Chi-square, df is degrees of freedom, and RMSEA is the root mean square error of approximation.*, **, and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.

1 .
Effects of Warming and Increased Precipitation on the Root Length Production of the Stipa breviflora Community

4. 3 .
Prediction of Root Production and Root Turnover in the Desert Steppe in the Future, Based on Historical Climate Change

Table 1 .
Two-way ANOVA of ANPP, root production, and root turnover of warming and increased precipitation.

Table 1 .
Two-way ANOVA of ANPP, root production, and root turnover of warmin increased precipitation.