*3.7. Effects of Precipitation Changes and Temperature on the Relationship between Grassland Vegetation Diversity, Biomass, and Soil Bacteria and Fungi Diversity*

Soil bacteria and fungi *α*- and *β*-diversity and ALB all promoted plant diversity, but RB had a limited function. The soil bacteria and fungi communities were able to promote the

decomposition of soil nutrients, offering more nutrients for plant to absorb and allowing more different kinds of plants to grow. This is consistent with previous research [25]. allowing more different kinds of plants to grow. This is consistent with previous research [25].

Soil bacteria and fungi *α*- and *β*-diversity and ALB all promoted plant diversity, but RB had a limited function. The soil bacteria and fungi communities were able to promote the decomposition of soil nutrients, offering more nutrients for plant to absorb and

soil temperature and moisture levels, which may further change the diversity of soil

*3.7. Effects of Precipitation Changes and Temperature on the Relationship between Grassland* 

*Plants* **2021**, *10*, x FOR PEER REVIEW 16 of 22

*Vegetation Diversity, Biomass, and Soil Bacteria and Fungi Diversity* 

#### **4. Materials and Methods 4. Materials and Methods**

bacteria and fungi communities.

#### *4.1. Study Site 4.1. Study Site*

The study area was located in the desert steppe of the Sidunzi ecological field station of Ningxia in China (37◦470 N 107◦250 E), which is on the southern edge of the Mu Us Sandy land and the yellow transition zone from the soil plateau to the Ordos platform (Figure 10). The natural conditions are relatively poor, characterized by drought, low rainfall, and strong winds and storms, and the region has typical temperate continental monsoon climate. Its annual average temperature is 8.1 ◦C, its monthly average temperature is −13.0–22.7 ◦C, its extreme maximum temperature was 34.9 ◦C, its extreme minimum temperature was −24.2 ◦C, its annual average frost-free period is 162 days, and its annual average precipitation is less than 300 mm. Its zonal soil structure is loose, and its soil fertility is low. Its zonal vegetation is typical of a desert steppe, and its dominant plants are Agropyron mogolicum Keng, Lespedeza potaninii Vass, and Polygala tenuifolia Willd. Due to the influence of climatic conditions and human activities, the grassland in the region has been degraded in large areas for a long time. The study area was located in the desert steppe of the Sidunzi ecological field station of Ningxia in China (37°47′ N 107°25′ E), which is on the southern edge of the Mu Us Sandy land and the yellow transition zone from the soil plateau to the Ordos platform (Figure 10). The natural conditions are relatively poor, characterized by drought, low rainfall, and strong winds and storms, and the region has typical temperate continental monsoon climate. Its annual average temperature is 8.1 °C, its monthly average temperature is −13.0–22.7 °C, its extreme maximum temperature was 34.9 °C, its extreme minimum temperature was −24.2 °C, its annual average frost-free period is 162 days, and its annual average precipitation is less than 300 mm. Its zonal soil structure is loose, and its soil fertility is low. Its zonal vegetation is typical of a desert steppe, and its dominant plants are Agropyron mogolicum Keng, Lespedeza potaninii Vass, and Polygala tenuifolia Willd. Due to the influence of climatic conditions and human activities, the grassland in the region has been degraded in large areas for a long time.

**Figure 10.** Location of the Sidunzi Village of Ningxia Observatory on the Loess Plateau. **Figure 10.** Location of the Sidunzi Village of Ningxia Observatory on the Loess Plateau.

#### *4.2. Experimental Design*

We completed all control experiment devices from June 2018 to March 2019. We started our experiment in May 2019 and collected soil and plant samples from July 2019.

According to meteorological monitoring of the study site from 1981 to 2017, its annual average precipitation, ground temperature, and air temperature all showed rising trends (Figure 11). Artificial rain-collecting greenhouses and sprinkler irrigation techniques were used to achieve 66% and 133% precipitation gradients and to ensure the precipitation treatment was kept within the 37-year average precipitation and fluctuation extremes. We

*Plants* **2021**, *10*, x FOR PEER REVIEW 17 of 22

*4.2. Experimental Design* 

established two temperature increase gradients to reflect the steady increases in ground temperature and air temperature recorded by meteorological monitoring. extremes. We established two temperature increase gradients to reflect the steady increases in ground temperature and air temperature recorded by meteorological monitoring.

We completed all control experiment devices from June 2018 to March 2019. We started our experiment in May 2019 and collected soil and plant samples from July 2019. According to meteorological monitoring of the study site from 1981 to 2017, its annual average precipitation, ground temperature, and air temperature all showed rising trends (Figure 11). Artificial rain-collecting greenhouses and sprinkler irrigation techniques were used to achieve 66% and 133% precipitation gradients and to ensure the precipitation treatment was kept within the 37-year average precipitation and fluctuation

**Figure 11.** The (**a**) Rainfall, (**b**) Ground temperature, and (**c**) Air temperature from 1981 to 2017. **Figure 11.** The (**a**) Rainfall, (**b**) Ground temperature, and (**c**) Air temperature from 1981 to 2017.

The designed rainout shelter was completed in November 2018, and these shelters were randomly built (Figure 12). The rainfall gradient was constructed with artificial shelters and sprinklers, and the rainout shelters were made of polycarbonate material, which can allow 90% of photosynthetic effective radiation to pass through it. A two-factor completely randomized experimental design was used based on rainfall and temperature factors. Five levels of rainfall 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 designed rainout shelter was completed in November 2018, and these shelters were randomly built (Figure 12). The rainfall gradient was constructed with artificial shelters and sprinklers, and the rainout shelters were made of polycarbonate material, which can allow 90% of photosynthetic effective radiation to pass through it. A two-factor completely randomized experimental design was used based on rainfall and temperature factors. Five levels of rainfall 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 and the interaction between the rainfall and the temperature, which was increased by about 1.5 ◦C with the OTC (Open-Top Chamber) in each plot [26]. The OTC was made of acrylic transparent board material, which can allow 90% of photosynthetic effective radiation to pass through it. The area of each plot was 6 × 6 m, and each treatment (n = 5) was repeated three times, for a total of 15 plots (temperature treatments are included in the precipitation treatments) (Figure 3). On the 15th and 30th of each month, R33 and R66 of the natural rainfall during the 1st–15th and the 16th–30th of the month, respectively, were collected from the actual rainfall and then evenly replenished to the plots containing R133 and R166 by a watering pot.

pot.

The temperature consisted of two levels: the actual temperature and the interaction between the rainfall and the temperature, which was increased by about 1.5 °C with the OTC (Open-Top Chamber) in each plot [26]. The OTC was made of acrylic transparent board material, which can allow 90% of photosynthetic effective radiation to pass through it. The area of each plot was 6 × 6 m, and each treatment (n = 5) was repeated three times, for a total of 15 plots (temperature treatments are included in the precipitation treatments) (Figure 3). On the 15th and 30th of each month, R33 and R66 of the natural rainfall during the 1st–15th and the 16th–30th of the month, respectively, were collected from the actual rainfall and then evenly replenished to the plots containing R133 and R166 by a watering

**Figure 12.** Rain shelter construction and Open-Top Chamber (OTC) arrangements of the subplots at the study sites. 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. **Figure 12.** Rain shelter construction and Open-Top Chamber (OTC) arrangements of the subplots at the study sites. 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.

#### *4.3. Collection of Soil Microorganism Samples*  In each plot, including the inner OTC, we collected 0–10 cm of soil from each sample *4.3. Collection of Soil Microorganism Samples*

plot. We removed the impurities in the soil, including plants, moss, visible roots, litter, and visible soil animals, and then wiped the sampler with alcohol-soaked cotton. After the alcohol had completely evaporated, we used the soil in the sample to soak the sampler. This step needed to be repeated each time the sample changed. Three points sampled from the same quadrant were mixed as one soil sample. We placed mixed soil into a 10 mL centrifuge tube and then transferred it to a −80 °C refrigerator for determination of soil microbes. We used the Operational Taxonomic Units (OTU) lever to determine the soil microorganism taxonomic group. The Operational Taxonomic Unit (OTU) is an operational definition used to classify groups of closely related individuals. In the context of numerical taxonomy, the most abundant sequence type was selected to represent each OUT [27]. In our research, OTUs are in the absence of traditional systems of biological classification (which are available for macroscopic organisms), pragmatic proxies for In each plot, including the inner OTC, we collected 0–10 cm of soil from each sample plot. We removed the impurities in the soil, including plants, moss, visible roots, litter, and visible soil animals, and then wiped the sampler with alcohol-soaked cotton. After the alcohol had completely evaporated, we used the soil in the sample to soak the sampler. This step needed to be repeated each time the sample changed. Three points sampled from the same quadrant were mixed as one soil sample. We placed mixed soil into a 10 mL centrifuge tube and then transferred it to a −80 ◦C refrigerator for determination of soil microbes. We used the Operational Taxonomic Units (OTU) lever to determine the soil microorganism taxonomic group. The Operational Taxonomic Unit (OTU) is an operational definition used to classify groups of closely related individuals. In the context of numerical taxonomy, the most abundant sequence type was selected to represent each OUT [27]. In our research, OTUs are in the absence of traditional systems of biological classification (which are available for macroscopic organisms), pragmatic proxies for "species" (microbial). For several years, OTUs have been the most commonly used units of diversity, especially when analyzing small subunit 16S for prokaryotes (as is the case of this work bacteria) or 18S (fungi) marker gene sequence datasets [28].

#### *4.4. Collection of Plant Samples*

The measurement of community *α*-diversity can be divided into the Shannon–Wiener diversity index, Margalef species richness index, Pielou evenness index, and Simpson dominance index. These four commonly used *α*-diversity indexes incorporate two measurements, the number (richness) of species, and the uniformity of species.

The richness index mainly measures the number of species within a certain spatial range to express the richness of organisms; the evenness index is a single statistic that combines the richness index and the evenness index; and the diversity index is based on the number of species to reflect a community's diversity, which can describe the disorder and uncertainty of an individual species. An increase in the number of species in the community represents an increase in the community's complexity. The greater the index value, the greater the amount of information contained in the community [9]. The importance value (IV) is calculated using the relative density, relative frequency, relative coverage, relative height, and relative biomass, according to the following formula:

Importance value (IV) = (Relative density + relative coverage + relative frequency + relative height + relative biomass)/5

We calculated plant diversity in terms of the number of species in each plot(s), the relative importance of the species in the plot (Pi), and the number of individuals in all species (N), according to the following formula:

Shannon–Wiener diversity index: H = − ∑ s <sup>i</sup>=<sup>1</sup> PilnPi Margalef species richness index: R = (S − 1)/log<sup>10</sup> N Pielou evenness index: E = H/lnS Simpson dominance index: C = 1 <sup>−</sup> <sup>∑</sup>(Pi)<sup>2</sup>

Dominant plant species: We measured the plant relative biomass, relative height, relative cover, relative frequency, and relative density to determine the importance values, and then used importance values to determine the dominant species.

Plant carbon, nitrogen, and phosphorus: Plant organic carbon and total carbon were measured with a total organic carbon (TOC) analyzer (CS Analysis Instrument, Naples, FL, USA), and total nitrogen and total phosphorus were measured with a HCLO4-H2SO<sup>4</sup> digestion-flow injection instrument (model Skalar-SAN++, Delft, The Netherlands).

Litter: We picked up the litter on the ground by hand in the sample squares that cut off the ground plants and carefully removed the fine soil particles attached to the litter and put them in the envelopes according to the sample squares. We dried samples at 65 ◦C to constant weight, weighed them, and recorded the dry weight data.

Plant height: The natural height of different plants in each sample box was measured 5 times, respectively. If there were not 5 plants, the plants outside the sample box were selected for measurement.

Plant coverage: The acupuncture method was used. A 1 m<sup>2</sup> square sample rope was placed on the ground and divided into an average of 100 grids. Plants were acupunctured in order every 10 cm from top to bottom and from left to right with a 2 mm needle. If the needle contacted with the plant, it was counted as 1, and it was not counted if there was no contact. The coverage of each plant in the sample was not more than 100. If two plants occurred simultaneously during acupuncture, the total coverage was reduced by 1, and if three plants occurred simultaneously, the total coverage was reduced by 2. (Note: Total coverage is the sum of coverage of each plant.)

Plant frequency: Rounds were thrown 10–15 times in the plot, and the total number of times that each plant appeared was the frequency of the plant. (Note: In each round thrown, as long as the plant appeared, regardless of the number of plants, the frequency was 1). The total frequency was the sum of the frequency of each plant.

Plant density: The kinds of plants and the number of times each plant appeared in the sample box were recorded.

Plant biomass: We measured plant biomass in a 1 m<sup>2</sup> quadrat that was randomly selected in each plot at the end of July 2019. We dug all plants in each plot out from the soil. We then cut the aboveground living plant and plant roots separately while sorting them according to species and placed them into respective envelopes. These species were then taken into the laboratory and dried at 65 ◦C in the oven for 48 h; the aboveground living plant biomass (ALB) and plant root biomass (RB) were then calculated.

### *4.5. Statistical Analysis*

We used repeated-measures ANOVA to examine the differences in the plant *α*-diversity index, the number of species and plants and the dominant species of the biomass, the organic carbon, total nitrogen, and total phosphorus of the plants and the dominant plant species under different precipitation levels and the interaction between precipitation and temperature by SPSS 21.0. *t*-tests were used to examine the differences in the plant *α*diversity index, the number of species and plants and the dominant species of biomass, and the organic carbon, total nitrogen, and total phosphorus of plants and dominant plant species under different temperatures by SPSS 21.0. The soil microorganism *β*-diversity was analyzed by principal component analysis (PCA) using R. We used principal components analysis (PCA) to examine the relationships between the grassland plant diversity, biomass, and soil bacteria and fungi *α-* and *β*-diversity by Origin 2021.

#### **5. Conclusions**

In this study, plant *α*-diversity in CK and TCK under the altered precipitation were significantly higher than other treatments. Under the interaction of the increasing precipitation and the rising temperature conditions, R166 promoted the number of species the most. Increasing precipitation was found to promote the growth of RB more than ALB, but the effect of rising temperatures on RB was not clear. The changing precipitation and increasing temperature factors, and the interaction of the two factors, all had no significant impact on the biomass, organic carbon, total nitrogen, and total phosphorus of plants. R166 promoted the ALB, RB, and total biomass of Agropyron mongolicum the most. The TR treatment promoted plant organic carbon, nitrogen, and phosphorus content more than R. In the fungi communities, under rising temperature, increasing precipitation promoted *α*-diversity, but *α*-diversity did not obviously vary in the bacteria communities. In the fungi communities, TCK promoted the most *β*-diversity, but in the bacteria communities, CK promoted the most *β*-diversity. Soil bacteria and fungi *α*- and *β*-diversity, and ALB promoted plant diversity the most.

**Author Contributions:** Conceptualization, J.L.; investigation, Y.Z., L.J., J.Z. and Y.W.; writing original draft preparation, Y.Z.; writing—review and editing, J.L.; project administration, Y.X., J.L. and H.M.; funding acquisition, Y.X. and J.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Ningxia key research and development program (2020BEG03046), the funder is Jian-Ping Li, and the Top Discipline Construction Project of Pratacultural Science (NXYLXK2017A01), the funder is Yingzhong Xie.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We thank the team from Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) for technical advice relating to our microbial analysis.

**Conflicts of Interest:** The authors declare no conflict of interest.
