*2.2. Plant α-Diversity 2.2. Plant α-Diversity*

The plant *α*-diversity had no significant differences among the different precipitation and temperature treatments (Table 1).

the same.


**Table 1.** Grassland plant *α*-diversity index under different precipitation and temperature treatments. F = 6.12 *p* > 0.05 F = 4.84 *p* > 0.05 F = 6.34 *p* > 0.05 F = 3.71 *p* > 0.05

The plant *α-*diversity had no significant differences among the different precipitation

*Plants* **2021**, *10*, x FOR PEER REVIEW 4 of 22

**Table 1.** Grassland plant *α-*diversity index under different precipitation and temperature treatments.  **Shannon–Wiener Pielou Margalef Simpson** 

and temperature treatments (Table 1)

Mean ± SE followed by lowercase letters in each column indicates significant differences between the variance percentage of precipitation, according to LSD test (*p* < 0.05). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pots: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pots: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were

#### *2.3. The Number of Species 2.3. The Number of Species*

Under the interaction of rising precipitation levels and increasing temperature conditions, the number of species was highest in CK, which was significantly higher than R33 (*p* < 0.05). With increased precipitation, the number of species was highest under R166, which was significantly higher than other precipitation treatments (*p* < 0.05). The difference between the natural temperature and the temperature increases was obvious (*p* < 0.05) (Figure 2). Under the interaction of rising precipitation levels and increasing temperature conditions, the number of species was highest in CK, which was significantly higher than R33 (*p* < 0.05). With increased precipitation, the number of species was highest under R166, which was significantly higher than other precipitation treatments (*p* < 0.05). The difference between the natural temperature and the temperature increases was obvious (*p*  < 0.05) (Figure 2).

**Figure 2.** *Cont*.

**Figure 2.** The number of species in the study sites. (**a**) The number of species under precipitation changing treatment (R); (**b**) The number of species under interaction of precipitation changing and temperature increasing treatment (TR). (**c**) The number of species under temperature increasing treatment (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), R33 was the first site of 33% precipitation, and other marks were the same. Values indicate the mean ± SE; different letters represent a significant difference according to LSD test (*p* < 0.05). \* represents a significant difference according to *t*-test (*p* < 0.05). **Figure 2.** The number of species in the study sites. (**a**) The number of species under precipitation changing treatment (R); (**b**) The number of species under interaction of precipitation changing and temperature increasing treatment (TR). (**c**) The number of species under temperature increasing treatment (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), R33 was the first site of 33% precipitation, and other marks were the same. Values indicate the mean ± SE; different letters represent a significant difference according to LSD test (*p* < 0.05). \* represents a significant difference according to *t*-test (*p* < 0.05). **Figure 2.** The number of species in the study sites. (**a**) The number of species under precipitation changing treatment (R); (**b**) The number of species under interaction of precipitation changing and temperature increasing treatment (TR). (**c**) The number of species under temperature increasing treatment (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), R33 was the first site of 33% precipitation, and other marks were the same. Values indicate the mean ± SE; different letters represent a significant difference according to LSD test (*p* < 0.05). \* represents a significant difference according to *t*-test (*p* < 0.05).

#### *2.4. The Biomass of Plants and Dominant Plant Species 2.4. The Biomass of Plants and Dominant Plant Species 2.4. The Biomass of Plants and Dominant Plant Species*

The RB was significantly higher than the ALB. The differences of the ALB were not significant under the interaction of the changed precipitation and increased temperature conditions (*p* < 0.05). The root biomass was highest in R166 but lowest in R33, and their difference were very significant *(p* < 0.05). The difference of the ALB under the natural temperature and the increased temperature were not significant, the same as the RB (*p* < 0.05) (Figure 3). The RB was significantly higher than the ALB. The differences of the ALB were not significant under the interaction of the changed precipitation and increased temperature conditions (*p* < 0.05). The root biomass was highest in R166 but lowest in R33, and their difference were very significant *(p* < 0.05). The difference of the ALB under the natural temperature and the increased temperature were not significant, the same as the RB (*p* < 0.05) (Figure 3). The RB was significantly higher than the ALB. The differences of the ALB were not significant under the interaction of the changed precipitation and increased temperature conditions (*p* < 0.05). The root biomass was highest in R166 but lowest in R33, and their difference were very significant *(p* < 0.05). The difference of the ALB under the natural temperature and the increased temperature were not significant, the same as the RB (*p* < 0.05) (Figure 3).

**Figure 3.** *Cont*.

**Figure 3.** Variations of aboveground plant living biomass (ALB) and plant root biomass (RB) of vegetation in the study sites. (**a**) Aboveground plant living biomass (ALB) under precipitation changing (R)and the interaction of the precipitation changing and temperature increasing (TR). (**b**) Root biomass (RB) under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing (TR). (**c**) Aboveground plant living biomass (ALB) under temperature increasing (T). (**d**) Root biomass (RB) under temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature, which was increased by about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 are the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a no significant difference according to *t*-test (*p* < 0.05). **Figure 3.** Variations of aboveground plant living biomass (ALB) and plant root biomass (RB) of vegetation in the study sites. (**a**) Aboveground plant living biomass (ALB) under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing (TR). (**b**) Root biomass (RB) under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing (TR). (**c**) Aboveground plant living biomass (ALB) under temperature increasing (T). (**d**) Root biomass (RB) under temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature, which was increased by about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 are the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a no significant difference according to *t*-test (*p* < 0.05).

When precipitation was increased, Agropyron mongolicum had the highest RB under R166, and had the lowest RB under R33, the RB difference between the precipitation gradients was significant (*p* < 0.05). The ALB of Agropyron mongolicum was highest under the R166, which was significantly higher than other precipitation treatments, the same as the total biomass of Agropyron mongolicum (*p* < 0.05). The RB, ALB, and total biomass of Lespedeza bicolor were highest under R33, which were significantly higher than other precipitation gradients (*p* < 0.05). The RB, ALB, and total biomass of Polygala tenuifolia were the highest under natural precipitation, which were significantly higher than other precipitation treatments (*p* < 0.05) This phenomenon with regards to the Agropyron mongolicum, Lespedeza bicolor, and Polygala tenuifolia under the warming and precipitation interaction was the same (*p* < 0.05). With the increases temperature, the differences of the RB, ALB, and total biomass of Agropyron mongolicum and Lespedeza bicolor were significant, but the temperature increases had no obvious impact on the Polygala tenuifolia *(p* < 0.05) (Figure 4). When precipitation was increased, Agropyron mongolicum had the highest RB under R166, and had the lowest RB under R33, the RB difference between the precipitation gradients was significant (*p* < 0.05). The ALB of Agropyron mongolicum was highest under the R166, which was significantly higher than other precipitation treatments, the same as the total biomass of Agropyron mongolicum (*p* < 0.05). The RB, ALB, and total biomass of Lespedeza bicolor were highest under R33, which were significantly higher than other precipitation gradients (*p* < 0.05). The RB, ALB, and total biomass of Polygala tenuifolia were the highest under natural precipitation, which were significantly higher than other precipitation treatments (*p* < 0.05) This phenomenon with regards to the Agropyron mongolicum, Lespedeza bicolor, and Polygala tenuifolia under the warming and precipitation interaction was the same (*p* < 0.05). With the increases temperature, the differences of the RB, ALB, and total biomass of Agropyron mongolicum and Lespedeza bicolor were significant, but the temperature increases had no obvious impact on the Polygala tenuifolia *(p* < 0.05) (Figure 4).

*Plants* **2021**, *10*, x FOR PEER REVIEW 7 of 22

**Figure 4.** Variations of aboveground plant living biomass (ALB) and plant root biomass (RB) of dominant species in the study sites. (**a**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Agropyron mongolicum under the precipitation changing and temperature increasing(TR); (**b**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Lespedeza bicolor under the precipitation changing and temperature increasing(TR); (**c**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Polygala tenuifolia under the precipitation changing and **Figure 4.** Variations of aboveground plant living biomass (ALB) and plant root biomass (RB) of dominant species in the study sites. (**a**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Agropyron mongolicum under the precipitation changing and temperature increasing(TR); (**b**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Lespedeza bicolor under the precipitation changing and temperature increasing(TR); (**c**) Aboveground plant

temperature increasing(TR); (**d**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Agropyron

living biomass (ALB) and plant root biomass (RB) of Polygala tenuifolia under the precipitation changing and temperature increasing(TR); (**d**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Agropyron mongolicum under the precipitation changing (R); (**e**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Lespedeza bicolor under the precipitation changing (R); (**f**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Polygala tenuifolia under the precipitation changing (R); (**g**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Agropyron mongolicum under the temperature increasing (T); (**h**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Lespedeza bicolor under the temperature increasing (T); (**i**) Aboveground plant living biomass (ALB) and plant root biomass (RB) of Polygala tenuifolia under the temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature, which was increased by about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE; different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a nonsignificant difference according to *t*-test (*p* < 0.05).

#### *2.5. The Organic Carbon, Total Nitrogen, and Total Phosphorus of Plants and Dominant Plant Species*

The differences in plant organic carbon, plant nitrogen, and plant phosphorus content were not significant under the changing precipitation condition and the interaction of the changing precipitation and increasing temperature conditions (*p* < 0.05), but the plant organic carbon, plant nitrogen, and plant phosphorus content under the precipitation condition changed less than under the interaction of the temperature and precipitation conditions. With the increases in temperature, the differences of the plant organic carbon, plant total nitrogen, and plant total phosphorus were also not obvious (*p* < 0.05) (Figure 5).

For the interaction of the temperature and precipitation conditions, plant organic carbon was the lowest at R166 for *Agropyron mongolicum*, which was significantly lower than other treatments (*p* < 0.05). Plant organic carbon was the highest at R66 for Lespedeza bicolor and was highest at R133 for Polygala tenuifolia, which were both significantly higher than other treatments (*p* < 0.05). Plant total nitrogen and plant total phosphorus were not significant among the different treatments (*p* < 0.05). The plant organic carbon, total nitrogen, and total phosphorus for *Agropyron mongolicum*, Lespedeza bicolor, and Polygala tenuifolia at the precipitation changed was lower. With the increased temperature, the differences of plant organic carbon, plant total nitrogen, and plant total phosphorus for *Agropyron mongolicum*, *Lespedeza bicolor*, and *Polygala tenuifolia* were not significant (*p* < 0.05) (Figure 6).

**Figure 5.** Percentages of plant organic carbon, plant total nitrogen, and total phosphorus of vegetation in the study sites. (**a**) Plant organic carbon under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing(TR); (**b**) Plant total nitrogen under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing(TR); (**c**) Plant total phosphorus under precipitation changing (R)and the interaction of the precipitation changing and temperature increasing(TR); (**d**) Plant organic carbon, plant total nitrogen, plant total phosphorus under temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 is the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a nonsignificant difference according to *t*-test. **Figure 5.** Percentages of plant organic carbon, plant total nitrogen, and total phosphorus of vegetation in the study sites. (**a**) Plant organic carbon under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing (TR); (**b**) Plant total nitrogen under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing (TR); (**c**) Plant total phosphorus under precipitation changing (R) and the interaction of the precipitation changing and temperature increasing(TR); (**d**) Plant organic carbon, plant total nitrogen, plant total phosphorus under temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 is the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a nonsignificant difference according to *t*-test.

For the interaction of the temperature and precipitation conditions, plant organic carbon was the lowest at R166 for *Agropyron mongolicum*, which was significantly lower than other treatments (*p* < 0.05). Plant organic carbon was the highest at R66 for Lespedeza bicolor and was highest at R133 for Polygala tenuifolia, which were both significantly higher than other treatments (*p* < 0.05). Plant total nitrogen and plant total phosphorus were not significant among the different treatments *(p* < 0.05*).* The plant organic carbon, total nitrogen, and total phosphorus for *Agropyron mongolicum*, Lespedeza bicolor, and Polygala tenuifolia at the precipitation changed was lower. With the increased

*Plants* **2021**, *10*, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/plants

significant (*p* < 0.05) (Figure 6).

first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 is the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). ns represents a nonsignificant difference according to *t*-test.

For the interaction of the temperature and precipitation conditions, plant organic carbon was the lowest at R166 for *Agropyron mongolicum*, which was significantly lower than other treatments (*p* < 0.05). Plant organic carbon was the highest at R66 for Lespedeza bicolor and was highest at R133 for Polygala tenuifolia, which were both significantly higher than other treatments (*p* < 0.05). Plant total nitrogen and plant total phosphorus were not significant among the different treatments *(p* < 0.05*).* The plant organic carbon, total nitrogen, and total phosphorus for *Agropyron mongolicum*, Lespedeza bicolor, and Polygala tenuifolia at the precipitation changed was lower. With the increased temperature, the differences of plant organic carbon, plant total nitrogen, and plant total phosphorus for *Agropyron mongolicum*, *Lespedeza bicolor*, and *Polygala tenuifolia* were not

**Figure 6.** Percentage of plant organic carbon, plant total nitrogen, and total phosphorus of dominant species in the study sites. (**a**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the interaction of the precipitation changing and temperature increasing (TR); (**b**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the interaction of the precipitation changing and temperature increasing(TR); (**c**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the interaction of the precipitation changing and temperature increasing(TR); (**d**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the precipitation changing (R); (**e**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the precipitation changing (R); (**f**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the precipitation changing (R); (**g**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the temperature increasing (T); (**h**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the temperature increasing (T); (**i**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased **Figure 6.** Percentage of plant organic carbon, plant total nitrogen, and total phosphorus of dominant species in the study sites. (**a**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the interaction of the precipitation changing and temperature increasing (TR); (**b**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the interaction of the precipitation changing and temperature increasing(TR); (**c**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the interaction of the precipitation changing and temperature increasing(TR); (**d**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the precipitation changing (R); (**e**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the precipitation changing (R); (**f**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the precipitation changing (R); (**g**) The organic carbon, total nitrogen, total phosphorus of Agropyron mongolicum under the temperature increasing (T); (**h**) The organic carbon, total nitrogen, total phosphorus of Lespedeza bicolor under the temperature increasing (T); (**i**) The organic carbon, total nitrogen, total phosphorus of Polygala tenuifolia under the temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall

precipitation increased, *α*-diversity first decreased and then increased.

precipitation conditions, the *α*-diversity did not show any obvious patterns.

In the fungi communities, *α*-diversity gradually decreased under the interaction of the increasing temperature and precipitation conditions. However, when only

In the bacteria communities, under the control of the increasing temperature and

In both the fungi and bacteria communities, *α*-diversity increased with increased

mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05).

*2.6. Soil Microorganism α-Diversity* 

temperature (Figure 7).

rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was

conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 are the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 are the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05). temperature increasing (T). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 are the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 are the same. Values indicate the mean ± SE, and different letters represent a significant difference according to LSD test (*p* < 0.05).

*Plants* **2021**, *10*, x FOR PEER REVIEW 3 of 21

#### *2.6. Soil Microorganism α-Diversity 2.6. Soil Microorganism α-Diversity*

In the fungi communities, *α*-diversity gradually decreased under the interaction of the increasing temperature and precipitation conditions. However, when only precipitation increased, *α*-diversity first decreased and then increased. In the fungi communities, *α*-diversity gradually decreased under the interaction of the increasing temperature and precipitation conditions. However, when only precipitation increased, *α*-diversity first decreased and then increased.

In the bacteria communities, under the control of the increasing temperature and precipitation conditions, the *α*-diversity did not show any obvious patterns. In the bacteria communities, under the control of the increasing temperature and precipitation conditions, the *α*-diversity did not show any obvious patterns.

In both the fungi and bacteria communities, *α*-diversity increased with increased temperature (Figure 7). In both the fungi and bacteria communities, *α*-diversity increased with increased temperature (Figure 7).

**Figure 7.** Soil microbial α*-*diversity of (**a**) fungi and (**b**) bacteria in the study sites by principal component analysis (PCA). Sobs index was the observed richness. Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. **Figure 7.** Soil microbial *α*-diversity of (**a**) fungi and (**b**) bacteria in the study sites by principal component analysis (PCA). Sobs index was the observed richness. Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same.

#### *2.7. Soil Bacteria and Fungi β-Diversity 2.7. Soil Bacteria and Fungi β-Diversity*

In the fungi communities, the distance between each sample point was the farthest under TCK, and the distance was 25124 according to PCA; therefore, the corresponding *β*-diversity was the highest under TCK. In the bacteria communities, the distance between each sample point was the farthest under CK; therefore, the corresponding *β-*diversity was the highest under CK, and the distance was 3010 according to PCA (Figure 8). In the fungi communities, the distance between each sample point was the farthest under TCK, and the distance was 25124 according to PCA; therefore, the corresponding *β*-diversity was the highest under TCK. In the bacteria communities, the distance between each sample point was the farthest under CK; therefore, the corresponding *β*-diversity was the highest under CK, and the distance was 3010 according to PCA (Figure 8).

*Plants* **2021**, *10*, x FOR PEER REVIEW 13 of 22

 **Figure 8.** Soil microorganism *β*-diversity of (**a**) fungi and (**b**) bacteria in the study sites by principal component analysis (PCA). Five levels of rainfall (R)were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 °C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 °C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same. **Figure 8.** Soil microorganism *β*-diversity of (**a**) fungi and (**b**) bacteria in the study sites by principal component analysis (PCA). Five levels of rainfall (R) were used: 33% (R33), 66% (R66), 100% (CK), 133% (R133), and 166% (R166) of the annual average. The first two rainfall conditions were obtained by using two rainout shelters with two manipulated rainfall doses: 97 mm (R33) and 194 mm (R66). For the three other rainfall conditions, we artificially increased rainfall in unsheltered plots using a watering pot: 295 mm (CK), 392 mm (R133), and 490 mm (R166). The temperature consisted of two levels: the actual temperature (CK) and the interaction between rainfall and the temperature, which was increased by about 2 ◦C (T) with the OTC (Open-Top Chamber) in each plot. TR33 was the first site of interaction between 33% precipitation (R33) and the temperature increase of about 2 ◦C (T), and the marks of TR66, TCK, TR133, TR166 were the same. R33 was the first site of 33% precipitation, and the marks of R66, CK, R133, R166 were the same.

#### *2.8. The Relationship between Grassland Plant Diversity, Biomass, Soil Bacteria, and Fungi αand β-Diversity 2.8. The Relationship between Grassland Plant Diversity, Biomass, Soil Bacteria, and Fungi α- and β-Diversity*

The Shannon–Wiener diversity index was positively correlated with the ALB, *α*diversity, and *β*-diversity but negatively correlated with the RB. The Pielou evenness index was positively correlated with *α*-diversity and *β*-diversity but negatively correlated with the ALB and RB. The Margalef species richness index was positively correlated with the RB and ALB. The Simpson dominance index was positively correlated with *α*-diversity and *β*-diversity but negatively correlated with the RB and ALB. The RB was greatly negatively correlated with *β*-diversity (Figure 9). The Shannon–Wiener diversity index was positively correlated with the ALB, *α*diversity, and *β*-diversity but negatively correlated with the RB. The Pielou evenness index was positively correlated with *α*-diversity and *β*-diversity but negatively correlated with the ALB and RB. The Margalef species richness index was positively correlated with the RB and ALB. The Simpson dominance index was positively correlated with *α*-diversity and *β*-diversity but negatively correlated with the RB and ALB. The RB was greatly negatively correlated with *β*-diversity (Figure 9).

*Plants* **2021**, *10*, x FOR PEER REVIEW 14 of 22

**Figure 9.** Principal components analysis (PCA) plots showing the influence of the Shannon–Wiener, Pielou, Margalef, and Simpson indexes, and the above-living biomass (ALB), root biomass (RB), *α-*diversity, and *β-*diversity, which represented effects of different temperature (recorded as CK and T) and variation in precipitation (recorded as R33, R66, CK, R133, **Figure 9.** Principal components analysis (PCA) plots showing the influence of the Shannon–Wiener, Pielou, Margalef, and Simpson indexes, and the above-living biomass (ALB), root biomass (RB), *α*-diversity, and *β*-diversity, which represented effects of different temperature (recorded as CK and T) and variation in precipitation (recorded as R33, R66, CK, R133, R166).

*3.1. Effects of Precipitation Changes and Temperature on Plant Main Value* 

#### **3. Discussion**

R166).

#### Under the changing precipitation condition and the interaction of the changing *3.1. Effects of Precipitation Changes and Temperature on Plant Main Value*

**3. Discussion** 

precipitation and the increasing temperature conditions, the main values of *Agropyron mongolicum*, *Lespedeza bicolor*, and *Polygala tenuifolia* were all higher than the other plants, so we made sure that the three plants were dominant plants; the reason might have been because the root system of *Lespedeza bicolor* forms vertical and horizontal networks in the soil layer, helping it to make full use of the water and nutrients therein. Furthermore, the proportion of woody and sclerenchyma cell tissue in Mongolia wheatgrass is large, and the roots of *Polygala tenuifolia* are strong and can absorb water better. Thus, these characteristics of the three plants give them better drought resistance than other species. *3.2. Effects of Precipitation Changes and Temperature on Plant α-Diversity*  Under the changing precipitation condition and the interaction of the changing precipitation and the increasing temperature conditions, the main values of *Agropyron mongolicum*, *Lespedeza bicolor*, and *Polygala tenuifolia* were all higher than the other plants, so we made sure that the three plants were dominant plants; the reason might have been because the root system of *Lespedeza bicolor* forms vertical and horizontal networks in the soil layer, helping it to make full use of the water and nutrients therein. Furthermore, the proportion of woody and sclerenchyma cell tissue in Mongolia wheatgrass is large, and the roots of *Polygala tenuifolia* are strong and can absorb water better. Thus, these characteristics of the three plants give them better drought resistance than other species.

#### Species diversity directly affects ecosystem function and stability, which is the foundation of human survival and development [19]. The distribution pattern of species *3.2. Effects of Precipitation Changes and Temperature on Plant α-Diversity*

diversity and its influencing factors have become the core problem of ecology and biogeography research [20]. The Shannon–Wiener and Simpson in CK and TCK are significantly higher than other treatments. This might have been because the sparse surface vegetation, loose soil Species diversity directly affects ecosystem function and stability, which is the foundation of human survival and development [19]. The distribution pattern of species diversity and its influencing factors have become the core problem of ecology and biogeography research [20].

structure, and low water-holding capacity of desert steppe. Precipitation directly affects the soil moisture content. The decrease of precipitation leads to the drought of topsoil and thus directly reduces the effective moisture content in soil. The change of soil moisture content indirectly affects soil nutrients and indirectly affects the absorption, transportation, and utilization of nutrients by plants by limiting the normal activities of rhizosphere microorganisms, resulting in the reduction of plant species. When The Shannon–Wiener and Simpson in CK and TCK are significantly higher than other treatments. This might have been because the sparse surface vegetation, loose soil structure, and low water-holding capacity of desert steppe. Precipitation directly affects the soil moisture content. The decrease of precipitation leads to the drought of topsoil and thus directly reduces the effective moisture content in soil. The change of soil moisture content indirectly affects soil nutrients and indirectly affects the absorption, transportation, and utilization of nutrients by plants by limiting the normal activities of rhizosphere microorganisms, resulting in the reduction of plant species. When precipitation increases, soil erosion occurs, and the reduction of carbon, nitrogen, and other nutrients in soil limits

the growth of plants. The distribution pattern of species diversity and its influencing factors have become the core problem of ecology and biogeography research.

#### *3.3. Effects of Changing Precipitation and Increasing Temperature on the Number of Species*

With increased precipitation, the number of species was highest under R166, possibly because plant roots need to absorb more water to grow in desert grasslands. This finding agrees with a previous study that found fine roots may have complex responses to the higher amount of precipitation predicted for the future [21].

#### *3.4. Effects of Changing Precipitation and Increasing Temperature on Plants and Dominant Species Biomass*

This study found that increased precipitation promoted the growth of RB more than ALB, but the effect of rising temperature on RB was not clear. When a plant is subjected to drought stress, it will reduce ALB and increase underground RB. Some studies have found that reducing drought stress caused by a lack of precipitation will prompt plant to allocate more biomass to the underground RB, so that the root system can better absorb water and nutrients in deep soil [22]. Plant biomass increases with rising precipitation, probably because increased precipitation can effectively supplement soil moisture and promote plant growth and development.

With increasing precipitation, R166 was found to promote the ALB and total biomass of *Agropyron mongolicum* the most, and R33 promoted the ALB, RB, and total biomass of *Lespedeza bicolor* the most. The ALB, RB, and total biomass of *Polygala tenuifolia* was highest under natural precipitation. Under rising temperatures, with increased precipitation, R166 was found to promote the total biomass of *Agropyron mongolicum* the most, and R33 promoted the total biomass of *Lespedeza bicolor* the most. *Polygala tenuifolia* had the highest biomass under natural precipitation, possibly because *Agropyron mongolicum* is a graminoid plant and needs to absorb more water than *Polygala tenuifolia* (*Leguminosae* plant) and *Lespedeza bicolor* (*Lespedeza bicolor* plant).
