Ecological adaptability and population growth tolerance characteristics of Carex cinerascens in response to water level changes in Poyang Lake, China

Water level conditions are the key factors that affect the growth and distribution of wetland plants. Using Carex cinerascens (C. cinerascens) as the study species, we employ indoor simulations and field surveys. Our results show that C. cinerascens can adapt to rhythmic changes in the water level through different adaptation strategies. Compared to that of the control group, plant growth was better with a 0–0.4 cm/d water level rate, and plant growth was in the 42–56 cm range to that a 1.0–1.4 cm/d water level rate. Furthermore, it was observed that 0–0.4 cm/d was the most suitable growth rate, with 0.6–1.0 cm/d and 0–32 cm being the ideal plant tolerance ranges, and increasing to 1.0–1.4 cm/d and 32–56 cm exceeds the plant tolerance threshold. In the middle and late period of the experiment (25–45 d), the ecological characteristics of the plants changed significantly. For example, the root-to-shoot ratio of the plant in the stable water level reached 26.1. In our field observations, plant biomass can be influenced by a variety of environmental factors. The frequency of the species was the largest at an elevation of 15 m, and the growth status of the dominant and companion species of C. cinerascens was weakened with an increase in soil moisture content. The suitable water content for C. cinerascens growth was 27.6–57.3%, the distribution elevation was 12.54–16.59 m, and the optimum elevation was 13.56–15.54 m. The study is expected to provide a reference for wetland ecology research and wetland protection and restoration, a theoretical reference for the coordination of water resource development and utilization of Poyang Lake and ecological protection of important lakes and wetlands, and an important scientific basis for wetland hydrologic regulation, ecological restoration and biodiversity conservation.

in the natural flow regime can also complicate the provision of ecosystem services and the sustainable use and management of standing water bodies.
In wetlands and lakes, water level fluctuations (WLFs) are very important factors that have a significant impact on the ecology, function and management of lakes 4,10,11 . The WLFs of lake ecosystems have significant temporal distribution characteristics and can be divided into two kinds of water level change modes: annual and interannual 12 , and these two variabilities can be greatly influenced by regional climatic conditions and human activities 13,14 . Therefore, numerous studies have focused on water pollution under the conditions of WLFs [15][16][17] . Some scholars are concerned about the impact of water level fluctuations on the overall structure, hydrological characteristics, and macrophytes of lakes and rivers [18][19][20][21][22] . However, the relative effects of WLFs on individual biological species and communities are less clear than the effects of other physical and chemical variables. There are few comprehensive studies on wetland plant individuals and populations under fluctuating water levels 23 .
Wetland plants are an important component of wetland ecosystems and the main primary producers. The distribution of wetland plants is affected by various environmental factors, such as topography, soil, water, and biology, and shows a certain distinct zonality 24 . This distribution shows significant temporal fluctuations in response to the hydrological regime by both direct and indirect effects 25,26 . Regarding direct effects, WLFs could not only cause a mixing disturbance 27 but also have a strong dilution effect on aquatic plant community dynamics 28,29 . With respect to indirect effects, water level reduction could increase sediment resuspension and turbidity, decrease light availability, and finally change the plant community structure 27,28 . At the same time, plants over the course of their life can form a series of morphological characteristics and population strategies to adapt to long-term periodic WLFs variability [30][31][32] . For example, in lakes, the difference in the spatial distribution of plants from lake shores to lake centers is a result of the differential adaptation and response of wetland plant populations to water conditions [33][34][35][36] . The hydrographic characteristics of wetlands are closely related to the compositional, structural and distributional characteristics of plant communities. The ecological amplitude of plants reflects the relationship between plants and the environment and the ability of plants to adapt to unfavorable environments. Therefore, the ecological breadth has an important influence on the distribution of plants 37 . According to Shelford's law of tolerance, organisms have tolerance ranges of upper and lower tolerance limits for ecological factors 38 . Previous studies have shown that the response of the ecological characteristics of wetland plants to hydrological factors typically conforms to a Gaussian distribution 39 . Therefore, the Gaussian model is the optimal model for calculating ecological amplitude 40 . Improving the knowledge base of the ecological adaptability and population characteristics of aquatic plants to changes in hydrological conditions is of great value for developing better predictions of the dynamics of aquatic communities and for the management of aquatic ecosystems in the context of climate change.
WLFs are also key ecological processes that influence plant population growth and distribution in the riparian zone 41,42 . The aquatic ecological environment changes with the pattern of water level fluctuations 43,44 . The relationship between WLFs and plant ecology has previously demonstrated that wetland plants are sensitive to large ranges of fluctuations and may in some cases be able to adapt to smaller and medium fluctuations to obtain more biomass [45][46][47] In some extreme cases, the individual growth and development and population formation and development of some species are completely dependent on water level fluctuations 48,49 . The study of plant responses to fluctuations is more focused on setting different water levels to explore their effects on plant population characteristics, species diversity and life history 19,[50][51][52][53][54] . Other studies analyzed the physiological and biochemical effects of water content or water depth gradients in different soils on plants 34,[55][56][57] . Part of the research focuses on the tolerance characteristics of plant population structure. For example, Juan-Ovejero et al. 58 found that the balance of tolerance to drought and saturation determined the temporal dynamics and vertical stratification of the soil invertebrate populations. Wang et al. 59 found that WLFs may potentially increase the vegetative spread of submerged macrophyte communities, and managing WLFs may be helpful for the restoration of submerged macrophyte communities in degraded wetlands. However, there are few studies on the growth tolerance characteristics of aquatic plant populations 59 . An experiment using seed bank material from two wetland studies has shown that moderate levels of water fluctuation could promote seed germination and seedling establishment, while intensive fluctuation could greatly restrict plant growth and distribution 60 . However, there have been fewer efforts that have focused on the ecological adaptability and population characteristics of C. cinerascens under different water environment conditions 23,61 ; in particular, the application of the Gaussian model is even rarer.
In this study, field observations and simulation experiments were combined. We examined the response of the growth process of C. cinerascens under WLFs and different field hydrological conditions to clarify the ecological adaptation mechanism and population characteristics in response to rhythmic hydrological changes in the Poyang Lake wetlands.

Materials and method
Study area and plant species. Poyang Lake (28°22′-29°45′N, 115°47′-116°45′E) is located in the middle and lower reaches of the Yangtze River in Jiangxi Province, China (Fig. 1). It is China's largest freshwater lake and an important internationally recognized wetland. It receives water inputs mainly from five rivers, the Fuhe, Ganjiang, Raohe, Xiushui, and Xinjiang rivers, as well as a number of other minor tributaries, and discharges into the Yangtze River from a narrow outlet in Hukou. WLFs are affected by water from the five rivers and the Yangtze River. The seasonal change in the water level of the lake is obvious and often experiences a water drop of up to 10 m 62 , showing a unique hydrologic rhythm that alternates between the wet season and the dry season 63,64 . In recent years, especially since 2000, the hydrological regularity of Poyang Lake has changed significantly with global climate changes and the disturbance caused by human activities, such as the construction and operation of the Three Gorges Water Control Project. The nature of these recent changes is that the low water period is advanced and prolonged, the high-water period is shortened and the overall water level is reduced. Additionally, www.nature.com/scientificreports/ the daily fluctuation range of the lake water level has increased 65,66 . The dramatic WLFs changes provide a wide range of beach extents and specific hydrological and soil conditions for the growth of wetland plants, so that different plant types present irregular distribution characteristics along elevation 67 and have a significant impact on the functions of this wetland ecosystem. C. cinerascens belongs to Carex of Cyperaceae, with rhizome and cluster perennial herbs, which are distributed in East China and central China. It is the most widely distributed dominant species in Poyang Lake, covering an area of 960 km 2 and accounting for 28.8% of the water surface area of Poyang Lake, and it plays an important role in the function of the Poyang Lake wetland ecosystem 68 .

Experimental design. Field investigations.
A field investigation of Baishazhou National Wetland Park was carried out in the dry season of Poyang Lake during November 2016. The analyses were carried out by considering the plant species, growth status, coverage, height and soil moisture status in each study area through the combination of a sample point method and a sample line method. Photos were taken of each sample point, GPS positioning was employed, plant samples were collected, comparisons were made with imaged indoor plants 69 for identification of plant species, and the elevations of various points were obtained according to longitude and latitude.
Simulation experiment. The laboratory experiment was carried out by the two-set basin method. The experimental site was the plant sunlight room of the Key Laboratory of Poyang Lake Wetland and Watershed Research Ministry of Education, Jiangxi Normal University, using a temperature of 22 ± 5 °C and natural light. On March 15, 2019, seedlings of C. cinerascens were collected from the typical distribution area in Nanji Wetland National Nature Reserve (28°55′45″N, 116°19′26″E) and then cultivated in a 17 cm × 18 cm pot for preculture, with ten plants per pot. The substrate was a meadow marsh soil collected from the Poyang Lake wetland (pH 5.36, organic matter content was 42 g/kg, and total nitrogen content was 19.6 g/kg). On April 10, plants with the same height (74.6 cm) were placed in the plexiglass bucket used for the experiment. Each glass barrel had a small hole with a diameter of 4 cm at a distance of 10 cm from the bottom, sealed with a rubber plug, which could be used to adjust the height of the water level in the glass barrel.
The test was from April 10 to May 20, 2019, for a total of 40 d. Three types of water level were used, the rising water level, the falling water level and the stable water level, which were set with seven water level gradients, 21 test groups and a 0-cm water level gradient (the water level does not pass the surface of the culture medium in the culture basin) control group. The test adopts a random area group setting with three repetitions for each  Data collection. In the field investigation portion of our study, the plant coverage was measured by the detection-visualized method, the community height was measured by tape measure, and the soil volume moisture content was measured and classified by the HH2 instrument according to the following categories: < 10% (dry), 10-30% (slightly wet), 30-50% (very wet), 50-70% (saturated), and > 70% (flooded) 70 . According to the soil volume moisture content, different weights were given to each type of soil moisture status, among which dry was 1, slightly wet was 2, very wet was 4, saturated was 6, and flooded was 8.5 70 .
Experimental indicators were measured once as a starting reference value before the water level fluctuated (10 April). Plant height was the highest plant in the same experimental treatment group measured by a tape measure every five days. The biomass of each group was measured, and the average biomass of each plant was calculated at 0, 25 and 45 days. The measured value after the end of the water level fluctuations of each test indicator (May 25) was taken as the final value.
The distribution of biomass was reflected by the root shoot ratio, and the accumulation of biomass was reflected by the absolute growth rate (AGR) and relative growth rate (RGR) in the biomass growth rate analysis. The calculation formulas are as follows: where B i and B i + 1 is the biomass of the Carex cinerascens population at T i and T i + 1 , respectively. Data analysis. The frequency of C. cinerascens in the elevation zone was analyzed using a Gaussian model to obtain the relationship between the growth of C· cinerascens and elevation 71 : Where y represents the index of the biological or ecological characteristics of the species, A + y 0 represents the maximum value of the corresponding index, x c represents the minimum value of certain environmental factors to plant species when the corresponding biological indicator reaches the maximum value, and w indicates the resistance of the species, which can also be described as the amplitude of the species. In general, suitable ecological amplitude is [x c -2w, x c + 2w], and optimum ecological amplitude is [x c -w, x c + w].
Experimental data were processed using Excel 2010 and Origin 2018 for data processing and drawing, the overview map of the study area was produced through arcgis 10.2. The statistical analysis of the measured data was carried out with SPSS 22.0. The least significant difference method was used to test the difference between the data. The repeated measurement variance method was used to analyze the relationship between C. cinerascens treatments in different time periods; before analysis, a sphere test was performed on repeated measurement data. If the test result was p > 0.05, the single-factor analysis of variance method was used; if the test result was p < 0.05, the multivariate analysis of variance method was used to process the data.

Results
Plant height. The variation trend of plant height was significantly different at different water levels during the experiment. When the experiment lasted for 20 days, the differences in plant height under different treatment intensities across three types of water levels were more obvious. In a stable water environment, for the water level range of 0-32 cm, the plant height increased with time, and the plant height increased with an increase in flooding depth. When the water level was greater than 40 cm, the plant height decreased with time. Throughout the stable water level experiment, the maximum increase in plant height occurred in the control group (56.46%), and the maximum decrease in plant height appeared at a flood depth of 56 cm (49.17%). In the WLFs test group, rates of water level change in the range of 0-0.6 cm/d induced an upward trend in plant height over time. For the range of 0.8-1.4 cm/d, the rising water level group first grew and then diminished, while the falling water level treatment group first decreased and then increased growth, and after 20 days, the plant height of the control group was higher than that of the other experimental groups. In both cases, the trend of plant height changed within a time range of 15-25 days. For the rising water level cases, the plant height increases in the 0.2 cm/d and 0.4 cm/d treatment groups were significantly higher than that of the control group ( p < 0.05), and the plant height of the 0.2 cm/d group was higher than that of the other experimental groups at 25-45 days. (Table 1)  www.nature.com/scientificreports/ groups showed a continuous increase; in the 40 cm flooding experiment, the biomass continued to increase. The aboveground biomass also decreased with time for a flood level of 40 cm. For submergence depths of 40 cm and 56 cm and a stable water level, the biomass trend first increased and then decreased. At the end of the experiment, the underground biomass of each group was significantly higher than the initial control value (0 d, 0 cm) (p < 0.05). The maximum increase in underground biomass was 29.67% (rising water level rate of 0.2 cm/d), and the minimum increase was 2.71% (falling water level rate of 1.0 cm/d). During the whole experimental period, the maximum aboveground biomass of the three water level types was 0.3896 g (rising water level, 0.2 cm/d), and the maximum total biomass was 1.1134 g (rising water level, 0.2 cm/d).
The change in time of the root-to-shoot ratio of C. cinerascens in different water environment types was roughly opposite the change in aboveground biomass. The root-to-shoot ratio of plants in the three water level environments was higher than that in other experimental groups. The root-to-shoot ratio increased continuously with time in the experimental treatment group, with a water level rate of change of 1.2-1.4 cm/d and a stable water level depth of 40-56 cm, and it decreased in the other experimental groups. In the later period of the experiment (25-45 days), the root-to-shoot ratio slightly increased or remained relatively stable. Over time, the largest reduction was in the control group, and the largest increase was in the 1.4 cm/d and 56 cm treatment groups. When the stable water level was 56 cm at the end of the experiment, the root-to-shoot ratio was the highest (26.10). (Figs. 2, 3, 4, 5).
Biomass accumulation. The biomass accumulation of C. cinerascens in different water environments highly varied in different experimental stages. The AGR of total biomass under each water environment type decreased with increasing experimental treatment intensity. The AGR of aboveground and underground biomass in the

Field distribution characteristics of the Carex population.
A total of 815 sample points and 56 species of plants were recorded in the field survey, of which 492 were dominant species, accounting for 60.37% of the sample points, and 67 were companion species, accounting for 8.22% (Fig. 1). The growth and distribution of the dominant species and companion species of C. cinerascens in the field are observed to have the same trend as the elevation changes, and they are mainly distributed in an area with an elevation of approximately 15 m. The number of dominant species in the range of 14-16 m elevation is as high as 344, and that of companion species is 47 (Fig. 8). The Gaussian regression analysis shows that a suitable distribution of C. cinerascens is in the elevation range of 12.54-16.59 m ( Table 2). The species community was mainly distributed in areas of soil moisture content less than 10% and soil moisture weights of 3.5-8.5 in the field environment. Conditions of 30% soil moisture as well as flood environments showed community growth that was better than that for conditions of 30-70% soil moisture, and there was no significant difference between the community coverage and height for dry, slightly wet and flooded conditions (p > 0.05). Conversely, for very wet and saturated conditions, the difference was significant (p < 0.05). Additionally, the growth of the companion species of C. cinerascens was similar to that of the dominant species in different soil moisture content environments (Figs. 9, 10).

Discussion
Adaptability of the growth of C. cinerascens to environmental water level changes. The growth conditions, such as plant height, stem length and leaf shape, of the aboveground sections of hygrophytes are the most direct response to changes in water level. To adapt to various conditions of WLFs, wetland plants usually make corresponding adjustments to maintain their own survival and growth 71,72 . The ability and strategy of plants to adapt to these environmental changes determine their proliferation and distribution range. An www.nature.com/scientificreports/ adjustment of biomass allocation is a common way for many wetland plants to adapt to changes in water level 73 . Plants living in changing habitats have adaptive strategies for the plasticity of biomass, and the results of this self-regulation are often in line with the predictions of the optimal allocation theory 74 , i.e., plants can utilize limited resources of light, nutrition and water by adjusting the biomass allocation of various organs to respond to changes in environmental conditions 75 . In the long-term adaptive evolution process, wetland plants have formed special survival strategies to reduce the impact of WLFs on plants. The most important thing is to respond to changes in the water environment through the fast-growing escape strategy and dormant, slow-growing tolerance strategy in the above-ground part 76,77 .
The results of our study determine that C. cinerascens has different survival strategies at different water levels. In water level environments of 1.2 cm/d, 1.4 cm/d and 40-56 cm, the oxygen supply and light in the water body are insufficient, photosynthesis is restricted, leaves wither and plants die. Furthermore, these conditions lead to a decrease in plant height, total biomass and aboveground biomass. The biomass is mainly concentrated in the underground components of the plant (Fig. 3). Zhang et al. 77 . found that the growth and photosynthesis of Carex schmidtii under drying treatment and reflooding treatments also decreased to different degrees. That is, the tolerance mechanism of the plant into a dormant state changes the life cycle of the life cycle strategy to www.nature.com/scientificreports/ respond to changes in the water environment to increase survival 78,79 . The growth and development process of plants for WLFs rates of 0.2 cm/d and 0.4 cm/d was significantly faster than that observed in the constant water level control group, since the water level was always lower than 8 cm and 16 cm, respectively, and the plants had time to adapt to each flooding depth. Thus, they could benefit from the resource allocation process before each water level change 80 . For a rising water level, C. cinerascens increased growth with an increase in water level; plant height, root-to-shoot ratio and biomass, exhibited varied trends over time (Figs. 2,3,4,5,6,7,8). The laboratory results, combined with field observations of plants and biomass in the wet and dry seasons, revealed that C. cinerascens can adopt an escape strategy for this intensity range and can adapt to changes in the water level environment through tolerance strategies. For a falling water level, the plant was affected by the initial water level height and the rate of WLFs. In the early and later periods of the experiment, the adaptive strategy was not utilized in the rising water level cases. Compared with rising water levels, plants with lower water levels at the same water level fluctuation rate have worse growth and development. Over time, the biomass of C. cinerascens began to gradually shift to the underground sections of the plant, resulting in a gradual slowing of plant height growth and the accumulation rate of aboveground biomass, while the accumulation of the underground biomass  In addition to hydrological factors, soil nutrients, temperature, precipitation and intraspecific and interspecific relationships of plants will also have an impact, and the ecological characteristics of wetland plants are the result of multiple factors 82 . In the indoor simulation experiment, the ecological indexes of C. cinerascens showed no obvious Gaussian distribution model under different water level conditions. This may be caused by the large difference between the water level gradient set in this experiment and the water level tolerance range of C. cinerascens. Gaussian regression is an effective method to study the distribution characteristics of plant populations along an environmental gradient 83 . In this study, the relationship between the population distribution and elevation of C. cinerascens in the wild environment is reflected.   www.nature.com/scientificreports/ The ecological amplitude of the water level is larger for flooding and drought conditions 26 . The growth and development of plants were severely inhibited at 42-56 cm, especially in the range of 20-45 days, as the total biomass decreased significantly, and the biomass accumulation was much lower than that in the first 20 days (Figs. 3,4,5,6,7,8). It was observed that plants can accumulate more biomass in a WLFs of 0-0.4 cm/d. For a WLFs of 0.6-1.0 cm/d, the indicators gradually began to show differences, indicating that plants were beginning to adapt to the changing water level for this range. However, field investigations showed that due to the uncertainty of the WLFs of Poyang Lake, the time of beach exposure varies, which makes the beach grow different plant communities at different elevations. However, due to the change in the hydrological situation of Poyang Lake in recent years, the growth range of C. cinerascens has changed from 15-17 m 84,85 to 11 m. This result is similar to Qi et al. 's 86 observation that water depth affects the development and expansion of Carex appendiculate.

Conclusion
There is an apparent threshold water level observed that allows better growth and development of C. cinerascens. For a stable water level, the tolerance range of plants was 0-32 cm, and the extreme growth tolerance was 32 cm. Under changing water level conditions, 0-0.4 cm/d was the optimal rate for healthy plant growth, and 0.6-1.0 cm/d was the fluctuation range of plant growth tolerance; 1.0-1.4 cm/d was over the tolerance threshold of plants, and the plant could not grow normally and complete its life cycle. In the field environment, the