Ecophysiological response of native and exotic salt marsh vegetation to waterlogging and salinity: Implications for the effects of sea-level rise

The ecophysiological characteristics of native Phragmites australis and exotic Spartina alterniflora grown under waterlogging and salinity were investigated to explore their adaptation potential to sea level rise. The seasonal course of phenotypic traits, photosynthetic activity and chlorophyll fluorescence parameters of P. australis did not change remarkably under shallow flooding, whereas these variables were sensitive to increasing salinity. Waterlogging exacerbated the negative effects of salinity on shoot growth and photosynthetic activity of P. australis, and the combined stresses led to an absence of tassel and reproductive organs. By contrast, S. alterniflora performed well under both stresses and showed an obvious adaptation of salt secretion with increasing salinity. Light salinity was the optimal condition for S. alterniflora, and the tassel growth, chlorophyll content and fluorescence characters under moderate stresses did not differ notably. The Na+ and Cl− concentrations in leaves of both species increased, and the K+ content decreased in response to salinity. Under moderate and high saline levels, the ion concentrations in S. alterniflora were maintained at relatively consistent levels with increased salt secretion. We expect the degradation of P. australis and further colonization of S. alterniflora under prolonged flooding and saltwater intrusion from sea level rise on the coastline of China.


Materials and Methods
Plant material and experimental setup. Integrated blocks of the plants with soil matrix were excavated from the largest salt marsh of the Chongming Dongtan (121°50′-122°05′E, 31°25′-31°28′N) located on the eastern fringe of Chongming Island in the Yangtze Estuary. In December (winter season) 2015, intact mesocosms consisting of organic soil monoliths (32 cm × 24 cm × 40 cm, a sufficient volume for root growth) with P. australis and S. alterniflora rhizomes were collected. The sampling sites of the two species were close to one another with similar soil properties, and the plant materials were collected at the same tidal line and had seedlings with similar life forms. Similar volumes of integrated soil monoliths were excavated according to the size of polyethylene containers, and the same soil was used to fill small gaps in the containers.
A total of 48 mesocosms were grown in a ventilated greenhouse with natural light and temperature conditions. A hose with a valve to control drainage was installed on the bottom of each polyethylene container. During January, the mesocosms were watered daily with fresh water to homogenize the soil salinity, which also provided a month of recovery from disturbance before experimental treatments. In early March, the mesocosms were fertilized with revised Hoagland's nutrient solution 26 once when the spring buds first appeared. Then, two flooding treatments, non-waterlogging (control group, the water level maintained at half of the container) and waterlogging (the water level maintained 50 mm above the soil surface), and four NaCl-salinity treatments, including freshwater (control group) and 5, 15 and 30 parts per thousand (ppt) salt, were established. The treatments for all containers (2 species × 2 waterlogging treatments × 4 salinity treatments × 3 replicates) were renewed biweekly, and fresh water was used to maintain the water level during the non-irrigation period to avoid excess salt accumulation.
Measurement of growth characteristics. From March to November, the stem height of three random shoots in each container was measured monthly. Total leaf area of the fully expanded leaves of a single plant was determined in situ using a grid paper counting method. In June, August and November, the aerial organs of three plants were harvested, and the biomass was measured. In November (later growing period), the tassels of three plants, with flowers and seeds, were harvested, and the biomass was measured. Harvested leaves, stems and tassels were dried separately in a forced-air oven at 60 °C for constant weight.
Measurement of photosynthetic parameters. In April, June, September and November, gas exchange measurements were conducted with a 2 cm × 3 cm standard leaf chamber using a portable steady-state photosynthesis system (Li-6400XT; Li-Cor Inc., Lincoln, NE, USA). Measurements were performed on the first fully expanded leaves under the flag leaf from 0800 to 1100 h on sunny and generally cloud-free days. Three plants in each container were measured. The net assimilation rates (A N ) at photosynthetic photon flux densities (PPFD) of 0, 20, 40, 60, 80, 100, 200, 400, 600, 800, 1200 and 1600 µmol m −2 s −1 were measured under a constant 1400 µmol mol −1 CO 2 . The light source was a red-blue LED light (Li-6400-02B; Li-Cor Inc., Nebraska, USA), and the CO 2 source for the measurements was the computer-controlled CO 2 mixing system supplied with the Li-6400XT equipment. Leaves were equilibrated at saturating PPFD before initiation of the light response curve. Sufficient time was allowed for the new PPFD to stabilize before logging the measurement (typically requiring 20 min or less). Dark respiration rates (R d ) of the shoots were measured after a period of at least 30 min in the dark. During the measurements, the temperature inside the leaf chamber was maintained at 25 ± 1 °C, and vapor pressure deficit was maintained at approximately 1.0 kPa. The relative humidity of the air in the leaf chamber was set to 60%. The shape of the average light response curve was modeled by fitting the data to a non-rectangular hyperbola equation 27 using a nonlinear least squares regression. The maximum area-based rate of photosynthesis (A max ) and the apparent quantum yield (Q α ) were used as the indicators of assimilation capacity (see Supplementary information).
SCIENtIfIC REPoRTS | (2018) 8:2441 | DOI:10.1038/s41598-017-18721-z Measurement of chlorophyll fluorescence parameters. In April and October, the chlorophyll fluorescence parameters were determined using an integrated leaf chamber fluorometer (Li-6400XT-40; Li-Cor Inc., Lincoln, NE, USA). Following the experimental protocol of Maxwell and Johnson 28 , the maximum (dark-adapted) photochemical efficiency of photosystem II (F v /F m ) and the instant (light-adapted) quantum yield of photosystem II (Φ PSII ) based on the PPFD-response curves were estimated (see Supplementary information).
Fluorescence was excited with a modulated red radiation of ca. 2 μmol m −2 s −1 by setting a pulse-width of 3 μs and a frequency of 20 kHz, and a saturating radiation pulse (0.8 s) of ca. 8000 μmol m −2 s −1 was provided. The minimum chlorophyll fluorescence (F 0 ) and the maximal chlorophyll fluorescence (F m ) of the closed photosystem II center were measured after 30 min of dark-adaptation. Subsequently, the leaves were continuously irradiated, and the fluorescence at the steady state (F s ) was thereafter recorded; a second saturating pulse ca. 8000 μmol m −2 s −1 was imposed to determine the maximal fluorescence of the light-adapted state (F m ′).
Estimation of salt secretion. During the growing period, salt secretion of S. alterniflora, as an adaptive response, was detected under salinity treatments. Because collecting the salt crystalloids secreted at the leaf surface was difficult, leaves were photographed (with scale bar), and the number of salt crystalloids was counted with CorelDRAW Graphics Suite X6. In the graphics interface, ten 0.5 cm × 0.5 cm grids were randomly chosen on each leaf for counting. In September, all the images were collected for both P. australis and S. alterniflora on the same day from 1500 to 1700 h, when the number of salt crystalloids was generally the highest.

Measurement of chlorophyll and ion concentrations.
Additionally, in September, the leaves undergoing photosynthetic measurements were destructively sampled and washed with deionized water to immediately determine the concentration of chlorophyll. Chl a and Chl b concentrations were measured according to the methodology described by Inskeep and Bloom 29 . Leaf samples (0.5 g fresh weight) were ground and extracted with 5 mL of 80% acetone, and then the supernatant was measured at 663 and 646 nm with a spectrophotometer (LAMBDA 950 UV/Vis/NIR; PerkinElmer Inc., MA, USA). The ratio of Chl a/b was also calculated.
Leaf samples were also dried and ground to a fine powder in a ball mill, and then the cations of sodium and potassium (Na + and K + ) and the anion of chlorine (Cl − ) were extracted with HNO 3 and AgNO 3 standard solutions, respectively. Concentrations of Na + and K + were determined using an atomic absorption spectrophotometer (Schimazu AA 6800; Schimazu Crop, Kyoto, Japan), and the concentration of Cl − was determined with a chloridometer (PXSJ-216; Leica Ltd., SH, China).
Statistical analyses. All data measured are presented as the mean ± standard deviation (S.D., n = 3) throughout the analyses. The results of growth, photosynthesis and chlorophyll fluorescence parameters and leaf characteristics for the native and exotic species were analyzed separately. The significance of the change and single-factor effects of waterlogging and salinity on the variables (except the PPFD-Φ PSII data) were analyzed with one-way analysis of variance (ANOVA) in conjunction with a post hoc Tukey HSD test, and the interactions were tested with two-way ANOVA. The level of statistical significance was set to P (probability) <0.05. Relationships between the salt secretion at the leaf surface and ion concentrations in the leaves were evaluated based on the optimal curve-fitting model with the highest coefficient of determination. All analyses of the variables were performed with the SPSS 23.0 statistical software package (SPSS Inc., Chicago, IL, USA).

Results
Plant growth. Regardless of saline treatment, the stem height and leaf area of P. australis under non-waterlogging were on average 15% and 16% higher than those under waterlogging, respectively, whereas in S. alterniflora, these differences were marginal throughout the growing period (Fig. 1). The aerial biomass and tassel mass of P. australis under non-waterlogging were on average 18% and 24% higher, respectively, than those under Salinities of 5, 15 and 30 ppt reduced the aerial biomass of P. australis by an average of 13%, 42% and 81%, respectively, compared with the control group throughout the measurement period, and the negative response to salinity was more pronounced during the later growing period than in earlier stages. At the end of the growing season, salinity significantly decreased the tassel mass of P. australis, and in some cases, tassels did not form under 15 and 30 ppt (Fig. 2). Regardless of flooding treatment, the maximum aerial biomass of S. alterniflora was obtained under salinity of 5 ppt, followed by non-salinity, 15 ppt and then 30 ppt.
An ANOVA test indicated a significant effect of waterlogging on the aerial biomass of P. australis only in August, but a significant effect of salinity on the stem height and leaf area of P. australis over the growing season (Table 1). A significant effect of salinity on aerial biomass and tassel mass of P. australis was found over June-November (Table 1), but the effect on S. alterniflora was not notable (except in August; Table 2). The combined treatments of waterlogging and high salinity (15 and 30 ppt) resulted in the lowest stem height, leaf area, aerial biomass and tassel mass of both P. australis and S. alterniflora, whereas no strong interactive effect of waterlogging × salinity was observed for the growth parameters.
Photosynthetic efficiency. Generally, the A max and Q α of P. australis increased from the early growing period and reached a peak in the summer season (e.g., June) and subsequently declined during the autumn season (Fig. 3). The peak A max in S. alterniflora was observed in early autumn (September), reflecting a longer growing period than that of P. australis. Regardless of saline treatment, the A max of P. australis under non-waterlogging was on average 12% higher than that under flooding over the measurement period, whereas the differences of A max and Q α of S. alterniflora were negligible. The photosynthetic parameters of P. australis were significantly affected by high salinity from summer to the end of the season ( Table 1). The A max and Q α were 11% and 5%, 23% and 16%, and 56% and 35% lower under 5, 15 and 30 ppt, respectively, than those of the control group throughout the measurement period, and the negative response to salinity was more pronounced during the later growing period than in earlier stages.
Regardless of flooding treatment, the saline treatments of 5 and 15 ppt did not change the A max and Q α of S. alterniflora dramatically compared with those of the control group, and 5 ppt salinity generally resulted in the maximum values during summer and autumn. The effect of salinity on the photosynthetic parameters of S. alterniflora was not notable (except Q α in November; Table 2). The lowest A max and Q α were measured under the combined treatments of waterlogging and high salinity for P. australis and S. alterniflora, whereas no significant interactive effect of waterlogging × salinity was detected for the photosynthetic parameters.
Chlorophyll fluorescence characteristics and chlorophyll content. Regardless of saline treatment, the chlorophyll fluorescence parameters of F 0 and F m were little affected by flooding treatment in either species at different measurement periods (Fig. 4). With the increase in salinity, the F 0 and F m of P. australis decreased and significantly reduced values were observed under 15 and 30 ppt. An ANOVA test indicated a significant effect of salinity on F 0 and F m of P. australis in both early and late growing periods (Table 1), whereas an effect on F 0 of S. alterniflora only occurred in a later period ( Table 2). The lowest values for F 0 and F m were found for P. australis and S. alterniflora under the combined treatments of waterlogging and high salinity, whereas the interactive effect of waterlogging × salinity was not significant. For both species, the values of F v /F m were slightly altered by waterlogging and saline treatments.
The responses of Φ PSII performance of both species to increasing PPFD showed a linear and then exponential decline to a low level at high PPFD (2000 μmol m −2 s −1 ) at earlier and later growing periods, respectively (Fig. 5). Waterlogging slightly affected (except when less than 30 ppt salinity) the Φ PSII of both species at different High salinity significantly decreased the Φ PSII of P. australis when PPFD was more than 1000 μmol m −2 s −1 at early growing periods and 500 μmol m −2 s −1 at the later growing period. Salinity at 5 ppt increased Φ PSII for S. alterniflora, whereas high salinity of 15 and 30 ppt slightly decreased values compared with those of the control group. At the later growing period, salinity significantly decreased the Φ PSII of S. alterniflora when PPFD was more than 1000 μmol m −2 s −1 .
As measured in early autumn (September), contents of Chl a and Chl b did not differ significantly between non-waterlogging and waterlogging treatments for both species (Table 3), regardless of saline treatment. Increasing salinity generally reduced the chlorophyll content of P. australis, with significantly lower values under salinity of 30 ppt than those in other treatments. However, the effect of salinity on chlorophyll content of S. alterniflora was not notable. The waterlogging and saline treatments did not change the ratios of Chl a/b for either species (with an exception for S. alterniflora).

Salt secretion and ion concentration.
Over the growing period, S. alterniflora secreted salt crystalloids from the leaves and stems, whereas this capacity was absent for P. australis (Fig. 6). As measured in September, the amount of salt crystalloids secreted at the leaf surface of S. alterniflora increased linearly with increasing salinity (Fig. 7a).
Concentrations of cations and the anion in P. australis leaves were higher than those in S. alterniflora leaves ( Table 3). Regardless of saline treatments, ion concentrations in the leaves of both species under waterlogging did not vary greatly compared with those in non-waterlogging. The concentrations of Na + and Cl − in P. australis and S. alterniflora leaves increased, whereas the K + content decreased in response to exposure to salinity, with contents significantly different from those of the control. The ion concentrations in P. australis leaves differed significantly between each salinity level. The differences of Na + and Cl − concentrations in S. alterniflora leaves between the highest two salinity levels (15 and 30 ppt) and the differences of K + concentration among all levels (5, 15 and 30 ppt) were not notable (Table 3).
In S. alterniflora leaves, both Na + and Cl − concentrations and salt secretion amount increased, and the K + content decreased with increasing salinity level. When the salinity level was higher than 15 ppt, the ion concentrations remained relatively consistent with the increased amount of salt secretion (Fig. 7b-d).

Discussion
P. australis and S. alterniflora are the typical native and exotic species, respectively, in many coastal salt marshes in China 21,22 . In the Yangtze Estuary, the tidal range and the intensity of saltwater intrusion are expected to increase with SLR projections 30 ; therefore, plant growth and survival and vegetation structure would be subject to the multiple stresses of prolonged tidal waterlogging and increased salinity.  We found that the seasonal shoot morphology and aerial biomass of P. australis were not significantly affected by a single waterlogging treatment, regardless of saline conditions. Additionally, the photosynthetic activity and chlorophyll fluorescence parameters of P. australis decreased slightly under waterlogging because of a marginal reduction in photosystem II activity, and the leaf characters of chlorophyll and ion concentrations did not differ significantly. The shallow flooding of 50 mm that was applied in this study provided an explanation for these results. In many coastal salt marshes, P. australis primarily occupies high marshes with low tidal water Tables 31-33 . Although grown under a long-term waterlogging (over the entire growing season), the growth and physiological characters of P. australis were not influenced remarkably by a low inundation projection. As reported by Mauchamp and Méthy 18 , only a long-term deep submersion resulted in an obvious decline of growth of P. australis, caused by severe damage to the photosynthetic apparatus based on fluorescence measurements. Regarding S. alterniflora, the aerial biomass was slightly higher under shallow flooding than under non-waterlogging throughout the growing season. Increased biomass with shallow flooding might be linked to the increase in photosynthesis and fluorescence parameters, reflecting that the water-saturated soil matrix might be optimal for the physiological requirements of S. alterniflora. In natural marshes, S. alterniflora generally colonizes the low marshes with a relatively high water level 33 . Previous studies show that biomass growth of S. alterniflora increases with rising sea level up to an optimum flooding frequency, showing greater biomass at intermediate than that at high elevation 11,34 . As reported by Naidoo et al. 35 , no substantial metabolic stress response of S. alterniflora is observed under waterlogging treatments. We also found that the roots of S. alterniflora were much finer and softer than those of P. australis, which contain abundant lignin (data not shown); therefore, the soft bulk soil saturated with water might benefit root development and growth of S. alterniflora.
However, the shoot development and biomass of P. australis were obviously sensitive to an increase in salinity. Therefore, this ecotype of the species in the Yangtze Estuary is salt-sensitive, and saline stress will be of great importance for plant growth and survival in coastal regions. The decreased biomass under saline treatment was primarily attributed to the decline in photosynthetic performance. High salinity suppresses the activity of carboxylation enzymes in chloroplasts and reduces osmotic pressure, resulting in limited photosynthetic activity [36][37][38] .  Under saline conditions, chlorophyll fluorescence parameters (except F v /F m ) of P. australis were reduced, indicating a reduction in photochemical efficiency in terms of light absorbed by chlorophyll associated with the photosystem II, reduced regeneration of ribulose-1,5-bisphosphate and the electron translocation rate [39][40][41] . Our study also showed that the effects of salinity on photosynthetic and fluorescence parameters were more pronounced during the later growing period than in earlier stages, revealing an accelerated degradation of the photosynthetic apparatus with aging. In this study, chlorophyll content of P. australis was not notably reduced under salinity of 5 and 15 ppt, which is consistent with previous reports for moderate saline conditions 42 . However, high salinity of 30 ppt significantly reduced the contents of Chl a and Chl b in P. australis. Based on the reviews by Ashraf and Harris 19 and Sudhir and Murthy 43 , the photosynthetic apparatus of most C 3 plants is impaired and the chloroplast pigments are degraded under salt stress. In adapting to saline environments, maintenance of intracellular ion homeostasis and osmotic adjustments are essential for plants 44 . The Na + and Cl − contents in P. australis leaves under all saline treatments were lower than those in S. alterniflora leaves, whereas the rate of increase of ion contents was much greater as the level of salinity increased from 5 to 30 ppt. Phragmites spp. generally present downward Na + and Cl − transport between shoot and root and ion accumulation in the shoot tissue in response to salinity 42,45 . Increasing salinity levels significantly decreased the K + concentrations in P. australis leaves. Barrett-Lennard and Shabala 46 suggested that such a change could be attributed to decreases in shoot relative growth rate. Moreover, P. australis does not have salt glands in the shoot tissues, and therefore, the capacity for salt secretion is absent, possibly resulting in much more accumulation of excess salt ions and limited uptake of K + in the leaves under high salinity levels. S. alterniflora showed better tolerance to high salinity than that of P. australis. Maximal seasonal biomass and photosynthetic parameters in S. alterniflora were often observed in the 5 ppt treatment, and the discrepancy of plant growth and photosynthetic activity under 15 ppt was not significantly different from non-salinity. Chlorophyll fluorescence parameters of S. alterniflora increased with a small increase in salinity (5 ppt), indicating that the electron transfer and turnover rate of the photosystem II might be promoted. These results are consistent with those of a field investigation of environmental gradients in the Yangtze Estuary showing that S. alterniflora grew better than the native species in salt marshes with high salinity 47 . The pattern of phenotypic plasticity of S. alterniflora might explain the successful invasion and strong competition with the native species in China 48 . A recent study 49 found that the growth of S. alterniflora under 4 ppt salinity was higher than that in a salt-free environment due to the regulation of antioxidant enzyme activities and the expression of key stress-induced proteins. Those researchers also predicted that S. alterniflora might tolerate salinity up to the concentration of local seawater (~12 ppt). S. alterniflora also maintained a relatively constant chlorophyll concentration across all salinity levels. The property of C 4 -type photosynthesis of S. alterniflora might contribute to some extent to tolerance to salt stress. In the C 4 pathway, salt ions might facilitate some biochemical processes in photosynthesis and water use efficiency under high salinity 16,50 . The greater salt tolerance of S. alterniflora than that of P. australis was due to the ability to use Na + for osmotic adjustment in the shoots 45 . Nevertheless, the biomass growth and biochemical parameters of S. alterniflora declined significantly under 30 ppt salinity based on the current measurements. Maricle et al. 51 reported that severe salinity decreases the gross rates of O 2 evolution and net rates of CO 2 uptake of S. alterniflora because of reduced stomatal conductance and increased photoinhibition. Compared with P. australis, an obvious adaptive mechanism of S. alterniflora to saline environments is salt secretion at the shoot surface with specialized salt glands 52 . With the increase in salinity from 0 to 30 ppt, the amount of salt crystalloids secreted from leaves increased sharply. Although the pattern of change of Na + and Cl − (increase) and K + (decrease) in S. alterniflora was the same as that of P. australis under saline treatments, the ion concentrations remained relatively consistent under high salinity levels in S. alterniflora. This adaptation mechanism is used to adjust osmotic potential and maintain smooth water inflow, in addition to maintaining the balance of K + and Na + in plants, to mitigate the adverse effects of saline stress 53,54 . Recent studies on salt secretion of halophytes indicate that various types of cations and anions can be secreted by salt glands in leaves, but Na + and Cl − are the dominant elements. By comparison, a low-affinity K + transporter inside the plasma membrane of salt glands maintains a high K + /Na + ratio in the cell cytosol to mitigate the salt stress 55 . Although statistical analysis did not show a significant interaction between waterlogging and saline stresses, the synergetic effects on P. australis were clear compared with S. alterniflora. Even shallow flooding exacerbated the negative effects of salt stress on biomass growth and photosynthetic activity. In particular, plants grown under flooding and 30 ppt salinity had the smallest shoots (stem height and leaf area) and the lowest biomass and photosynthetic capacity compared with those of plants grown under a single stress. The combined hypoxia and salinity are hypothesized to impair the photosynthetic apparatus and disrupt the ion homeostasis between shoots and roots 56,57 . Furthermore, this study showed that the tassel mass with reproductive organs of P. australis was significantly reduced under salinity. The combined stresses of waterlogging and salinity also led to an absence of tassel development under 15 and 30 ppt salinity; thus, sexual propagation would clearly be blocked during the next generation. Further monitoring of the developmental response mechanisms (e.g., asexual reproduction by rhizome tissue) and growth under consecutive stresses is required.
Vegetation zonation and spatial patterns are determined by species-specific adaptations to environmental gradients of water level, sediment type and salinity in estuarine marshes 58,59 . With SLR projections, some species would be excluded from the low and middle tidal zones because of weak tolerance to prolonged flooding and higher salinity, and interspecies competition would also lead to a shift of vegetation zonation under environmental stresses 60 . We found that, in addition to phenotypic characteristics, the level of tolerance of photosynthetic performance and biochemical characters to environmental stresses was higher for S. alterniflora than that for P. australis. Based on these differences, we predict that the native P. australis may become more vulnerable to increased flooding and saltwater intrusion, whereas these stresses may favor the exotic S. alterniflora. Therefore, our study supports the hypothesis that degradation of native P. australis and further colonization of exotic S. alterniflora might be driven by a combination of increased flooding and high saltwater loads under projected SLR conditions on the Chinese coastline.