Abstract
To explore whether grazing-induced legacy effects on plants could benefit plants adaptation to drought.
A water-controlled experiment was conducted in the greenhouse, which with Agropyron cristatum and Carex korshinskyi collected from free-grazing and enclosed plots on a typical grassland in Inner Mongolia.
We found that A. cristatum and C. korshinskyi collected from the free-grazing plot were less affected by drought in terms of ramet biomass, ramet number and total biomass than those collected from the enclosed plot. The enhanced adaptation to drought for plants collected from the free-grazing plot should partly be ascribed to the larger root biomass allocation plasticity under drought treatment. Our findings suggest that grazing management can be used to improve the adaptation of grassland plants to climate change.
摘要
为探索植物的放牧遗留效应是否有利于天然草原生态系统应对干旱环境，我们采集了 内蒙古典型草原自由放牧和多年围封样地内的冰草(Agropyron cristatum)和黄囊苔草(Carex korshinskyi)幼苗 进行了温室控水试验。研究结果表明，干旱处理对自由放牧样地采集的冰草和黄囊苔草子株生物量、子株数量和总生物量的影响较小；自由放牧区的冰草与黄囊苔草较强的干旱适应性可部分由干旱处理下较大的根系分配可塑性来解释。本研究结果表明合理放牧是天然草原适应气候变化的潜在管理办法之一。
INTRODUCTION
Climate change has attracted great attention from many ecologists and various organizations (Christensen et al. 2004). By the end of the 21st century, the global average temperature is projected to increase 1.8–4.0 °C, while the change of global average precipitation will show a tremendous spatial heterogeneity, from increasing 50% to decreasing 50% in different areas (Li et al. 2018; Solomon 2007). For the typical steppe in Inner Mongolia, there will be a decreasing trend in precipitation (especially in summer) and a greater increasing trend in temperature than in other areas (Mohammat et al. 2013; Solomon 2007; Wang et al. 2008). These changes will lead to increased drought stress on grassland plants in this region (Dai et al. 2004; Piao et al. 2011).
Grazing is the most common management method of grassland. Meanwhile, grazing could mediate the effects of drought on grasslands (Li et al. 2018). Conceptual and simulation models predict that synergistic interactions between drought and grazing may accelerate grassland degradation processes (Lohmann et al. 2012; Martin et al. 2014). However, grazing-induced effects of drought on grasslands may be influenced by grassland type, climatic condition, grazing intensity and grazing pattern. For example, while the effects of climate treatments were largely similar in the grazed and ungrazed plots in Tost, they were larger in the ungrazed plots in Spiti (Kohli et al. 2020).
Generally, grazing influences the effects of drought on grassland through soil water content, species diversity and species composition (Grime et al. 2000; Lohmann et al. 2012). Grazing-induced legacy effects on plants—a phenomenon where offsprings from either long-term enclosed or free-grazing plots show significant phenotypic differences under the same growing environment—may play an important role in mediating drought effect on grasslands (Guo et al. 2020). Firstly, compared with progenies in an enclosed area, grazing-induced more intense drought could prepare progenies to resist drought via stress memory (Tombesi et al. 2018; Walter et al. 2013). Secondly, some herbivory-resistant plant traits, such as small stature, rapid growth and belowground nutrients reserve, could enhance the adaption of grass plants to drought (Augustine and McNaughton 1998; Coughenour 1985; Veldhuis et al. 2014). However, there has been no study to explore whether grazing-induced legacy effects on plants could enhance the adaption of progeny plants to drought stress.
Agropyron cristatum and Carex korshinskyi are important components of vegetation community in the typical grasslands of Inner Mongolia, China (Yang et al. 2009). The dominance of both species were increased significantly under grazing disturbance. Furthermore, according to our observation, especially in the free-grazing plot, the density and frequency of A. cristatum and C. korshinskyi were higher than those of Leymus chinensis and Stipa chinensis, which are the dominant species in the undegraded typical grasslands. Therefore, we selected A. cristatum and C. korshinskyi to explore whether the grazing-induced legacies on plants benefit for grassland plants to drought through a water-controlled experiment in the greenhouse.
MATERIALS AND METHODS
Materials
Materials were collected from free-grazing and enclosed plots at a typical steppe grassland located at the Inner Mongolia Grassland Ecosystem Research Station (43°38′ N, 116°42′ E). The free-grazing and enclosed plots were adjacent and separated by a fence since 1983 at the sampling sites. The free-grazing plot has been overgrazed with sheep for more than 50 years at a stocking rate of ~3 sheep units per hectare while ~1.5 sheep units per hectare was the appropriate grazing capacity recommended by the local government.
Early reports (Xiu et al. 2014; Zhang et al. 2005) conducted at the same site have indicated that grazing induced few effects on plant genetic structure. We conduct this study at the population level as Didiano et al. did (2014). We sampled 100 A. cristatum and C. korshinskyi offsprings with similar status from the entire area of each plot at >20 m intervals except for the margin, respectively. Each offspring was cultivated in a pot (20 cm × 15 cm) in the greenhouse. The pot contained 3 kg of soil collected from the sampling site.
Experimental design
Seven days after being transplanted to the greenhouse, all offsprings recovered and the experiment was conducted. The study was a 2 × 2 full factorial design consisting of two factors imposed to A. cristatum and C. korshinskyi, respectively. The first factor was the source of the material, i.e. NG and OG. NG indicates offsprings collected from the enclosed plot while OG indicates offsprings collected from the free-grazing plot. The second factor was drought treatment, i.e. CK and DT. CK indicates the control treatment while DT indicates the drought treatment. In NG or OG treatment for each species, 50 pots of offsprings with similar growth status were selected to conduct the experiment. Twenty-five pots were assigned to CK and the other 25 pots were allotted to DT. All offsprings were watered using the weighting method every evening, and CK was watered to 40% content while DT was watered to 5% content.
Measurements
After counting the ramet number in each pot, the tallest ramet was selected for the measurement of ramet biomass. The tallest ramet and all other ramets were cut with scissors and put into an envelope, respectively. The total biomass for the material in each pot was the sum of the tallest ramet biomass, the other ramet biomass and the root biomass (for C. korshinskyi, the total biomass also consists of the rhizome biomass). The root biomass allocation was estimated using root biomass divided by total biomass. Finally, the pots were broken, and the belowground part of the plant and the soil were transferred into a 0.04 mm mesh nylon net, washed with water and enclosed in an envelope. All envelopes with materials (i.e. the tallest ramet, all other ramets and plant underground part) were oven-dried at 65 °C for 48 h. Since C. korshinskyi belowground part consists of root and rhizome, we weighed the root and rhizome separately.
Statistical analysis
Data were first subjected to the normal distribution and homogeneous invariance test. Logarithmic transformation was applied for data transformation to attain normality and homoscedasticity where necessary. For A. cristatum and C. korshinskyi, one-way analysis of variance (ANOVA) or Kruskal–Wallis test was used to analyze the effects of material source and drought treatment on ramet biomass, ramet number, total biomass and root biomass allocation. Two-way ANOVA, plastic index (PI) and Δ under drought treatment were used to analyze the different responses to drought treatment between NG and OG in terms of the ramet biomass, ramet number, total biomass and root biomass allocation. Indicators that did not conform to the normality or homoscedasticity were not included in the two-way ANOVA analysis. PI and Δ refer to the relative and absolute change to the drought treatment, respectively, and were calculated using the following formula (Valladares et al. 2000):
RESULTS
For A. cristatum under CK and DT treatment, there were no significant differences between NG and OG in ramet biomass and ramet number. For the drought treatment, the PI for ramet biomass and ramet number of OG were 44.80% and 80.17%, while the Δ were 0.04 and 22.90. For NG, the PI for ramet biomass and ramet number were 55.10% and 83.68%, while the Δ were 0.06 and 29.40 (Fig. 1a and b). Under CK, the total biomass of NG increased compared with OG (P < 0.05) while no difference was found between NG and OG under DT. There was a significant interactive effect of material source and drought treatment on total biomass (P = 0.004, Fig. 1c).
Responses of ramet biomass, ramet number and total biomass of A. cristatum (a–c) and C. korshinskyi (d–f) to drought treatment. NG indicates offsprings collected from the enclosed plot while OG indicates offsprings collected from the free-grazing plot. CK indicates the control treatment while DT indicates the drought treatment. Different capital letters represent significant differences between CK and DT. Different lowercase letters represent significant differences between NG and OG. PI and Δ indicate the PI and absolute variation to drought treatment, respectively. The PG×D above each subgraph indicates the interaction between drought treatment and materials source, while “NO” implies that the interaction effect was not estimated because of heteroscedasticity or lack of normality in the dataset.
For C. korshinskyi, NG accumulated more total biomass than OG under CK while OG showed larger total biomass than NG under DT. There were no significant differences between NG and OG in ramet number under CK and DT and in ramet biomass under DT. The PI for NG and OG in response to drought treatment were 70.43% and 62.02%, and the Δ were 6.85 and 4.8. Material source and drought treatment had significant interactive effects on ramet biomass and total biomass (Fig. 1d–f).
Both A. cristatum and C. korshinskyi allocated more biomass to the root when experiencing drought. For A. cristatum, no significant difference was observed between NG and OG in root biomass allocation under CK while OG allocated more biomass to root than NG under DT. For C. korshinskyi, there were no significant differences in root biomass allocation between NG and OG under CK and DT treatment. The PI and Δ in root biomass allocation under drought condition were −21.49% and −8.15% for OG, and −9.76% and −4.01% for NG (Fig. 2).
Responses of root biomass allocation of A. cristatum (a) and C. korshinskyi (b) to drought treatment. All parameters have the same meaning as those in Fig. 1.
DISCUSSION
Numerous studies have shown that the interactive effects between grazing and drought could accelerate grassland degradation (Lohmann et al. 2012; Qin et al. 2019). This may be attributed to increased evaporation and reduced soil water holding capacity under grazing disturbance (Keblawy 2016; Ritchie 2014). Our results indicated that grazing-induced legacy effects on A. cristatum and C. korshinskyi could enhance the adaptability of both species to drought. Thus, appropriate grazing pattern and intensity (e.g. light grazing, rotational grazing, etc.), which benefit to conserve soil moisture, may enhance the adaptation to climate change for grasslands by the grazing-induced legacy effects on plants.
According to the functional equilibrium theory, more biomass is allocated to the root for water absorption from soil under drought condition (Poorter et al. 2012). Many studies have provided evidence that plants adapt to drought by increasing root biomass allocation (Markesteijn and Poorter 2009; Quezada and Gianoli 2010). Our results also indicate that the root biomass allocations of A. cristatum and C. korshinskyi increased under the drought treatment. Furthermore, A. cristatum and C. korshinskyi from the free-grazing plot exhibited a greater increase in root biomass allocation under drought conditions than those from the enclosed plot. Therefore, the strong drought adaptation of A. cristatum and C. korshinskyi from the free-grazing plot could be ascribed to the extent of the increase in root biomass allocation.
Considering the little impact of grazing on population genetic structure (Jamnadass et al. 2006; Mengli et al. 2005), we speculate that the enhanced drought adaptation induced by grazing should be attributed to epigenetic pathways, rather than natural evolution. Grazing-induced drought in grasslands may play an important role in enhancing plant drought adaptation via epigenetic regulation, which is called stress memory (Tombesi et al. 2018; Walter et al. 2011). On the other hand, herbivorous-induced transgenerational effects on progenies to cope with herbivores may also contribute to the increased drought adaptation, which is called the cross stress memory (Munné-Bosch and Alegre 2013; Walter et al. 2013). However, to better understand how grazing disturbance increases plant adaptability to drought (herbivorous- or drought-induced, natural evolution or epigenetic regulations), warrants further studies under the cooperation between ecologists and biologists.
Although we found that the grazing-induced legacy effects on plants may be beneficial for grassland to cope with drought in a greenhouse experiment, more in situ tests are needed to validate this conclusion. On the other hand, the effect of grazing on the responses of grassland to drought is a comprehensive effect of multiple pathways. Thus, what grazing pattern and grazing intensity enhance the adaptation to drought for grasslands requires future studies. Furthermore, because population regeneration on grassland mainly relies on asexual reproduction and because it is difficult to collect enough seeds for this experiment, the materials used in our study were asexual offsprings. However, seed reproduction is also an important way in population regeneration for many species in grasslands. The asexual offsprings and seedlings may show different responses to grazing-induced legacy effects. Therefore, whether grazing-induced legacy effects enhance the drought adaptability of seedlings need further studies.
Funding
This research was supported by the Inner Mongolia Science and Technology project (201802081), the National Natural Science Foundation of China (32071882) and the Inner Mongolia Natural Science Foundation (2020MS03070).
Acknowledgements
We are sincerely grateful to the Inner Mongolia Grassland Ecosystem Research Station (IMGERS) of the Chinese Academy of Sciences and the National Agricultural Experimental Station for Soil Quality, Hohhot (NAES042SQ20) to provide the experiment sites for collecting the materials and conducting greenhouse experiment.
Conflict of interest statement. The authors declare that they have no conflict of interest.
REFERENCES
Author notes
These authors contributed equally to this work.
