Ecoengineering Solutions for the Impairment of Spreading and Growth of Invasive Spartina patens in Mediterranean Salt Marshes

The invasion of natural communities by non-indigenous species represents one of the most serious threats to biodiversity. Understanding the ecophysiology of invasive species can provide insights into potential physiological handicaps relative to native species. By doing so, we can leverage the development of ecoengineering solutions for the removal of non-indigenous species, preferably using non-chemical methods. Spartina patens is a known invasive species of cordgrass aggressively proliferating in Mediterranean salt marshes, producing impenetrable monospecific stands. As its occurrence is delimited by the upper high tide water level, we hypothesized that S. patens is intolerant to waterlogging. Therefore, we developed a field experiment where strands of S. patens were kept waterlogged over the entire tidal cycle for 30 days. At the end of the experimental period, plants in the trial plots exhibited severe stress symptoms at different physiological levels compared with control plots (no intervention). At the photobiological level, intervened plants exhibited lower efficiency in producing chemical energy from light, whilst at the biochemical level waterlogging impaired the antioxidant system and increased lipid peroxidation products. Furthermore, the application of chlorophyll a pulse amplitude modulated (PAM) fluorometry, a non-invasive technique, allowed us to evaluate the effectiveness of the implemented measures, being the tool that provided the best separation between the control and intervened population. Considering the physiological traits observed here, ecoengineering solutions based on increased waterlogging of S. patens stands, can be a low-cost and efficient measure to reduce the spreading and growth of this invasive species in the Mediterranean and other salt marshes worldwide with little disturbance.

The invasion of natural communities by non-indigenous species represents one of the most serious threats to biodiversity. Understanding the ecophysiology of invasive species can provide insights into potential physiological handicaps relative to native species. By doing so, we can leverage the development of ecoengineering solutions for the removal of non-indigenous species, preferably using non-chemical methods. Spartina patens is a known invasive species of cordgrass aggressively proliferating in Mediterranean salt marshes, producing impenetrable monospecific stands. As its occurrence is delimited by the upper high tide water level, we hypothesized that S. patens is intolerant to waterlogging. Therefore, we developed a field experiment where strands of S. patens were kept waterlogged over the entire tidal cycle for 30 days. At the end of the experimental period, plants in the trial plots exhibited severe stress symptoms at different physiological levels compared with control plots (no intervention). At the photobiological level, intervened plants exhibited lower efficiency in producing chemical energy from light, whilst at the biochemical level waterlogging impaired the antioxidant system and increased lipid peroxidation products. Furthermore, the application of chlorophyll a pulse amplitude modulated (PAM) fluorometry, a non-invasive technique, allowed us to evaluate the effectiveness of the implemented measures, being the tool that provided the best separation between the control and intervened population. Considering the physiological traits observed here, ecoengineering solutions INTRODUCTION Salt marshes provide a wide range of ecological services, including nursery habitats for many animals, protection against coastal erosion, water purification, having the considerable capacity to store and sequester carbon, and are key players in the ecosystem natural remediation capacity Teixeira et al., 2014;Duarte et al., 2018Duarte et al., , 2021, being these services valuated in several millions of euros per year (Duarte et al., 2021). Being preferred locations for human settlement and a profusion of anthropogenic activities, coastal, and transitional areas have been severely impacted in their health and functioning worldwide. Habitat loss and degradation, climate change and the introduction of invasive species are amongst the major threats to salt marsh ecosystems (Duarte et al., 2015Repolho et al., 2017;Pérez-Romero et al., 2018). Therefore, in the context of increasing degradation rates, the need to restore salt marsh ecosystems has been recognized as a priority by managers, scientists, and general society. This is well-emphasized in the EU Biodiversity Strategy for 2030 and its EU Nature Restoration Plan (EU, 2020) and reinforced by the UN Decade on Ecosystem Restoration (UN, 2020).
crates (Hultén, 1958). Therefore, although being present for some time along the Western Mediterranean coasts, it had not been recorded on the Eastern Iberian coast until recently (Baumel et al., 2016).
Controlling invasive species is relevant for restoration efforts, and several approaches have been developed to address this issue, ranging from herbicide application to mowing and physical removal (Kerr et al., 2016). While the first appears as a simple and cost-effective method (Major et al., 2003;Patten et al., 2017), the long-term effects of herbicides may have serious impacts on the ecosystem (Cruz de Carvalho et al., 2020b). As for the other two methods, they will only prevent further spread needing to be constantly applied and being very time-and cost-consuming (Hedge et al., 2003). Therefore, we considered an alternative method underpinned on the physiology of the species, in particular the fact that S. patens has a low tolerance to waterlogging, due to poor aeration of the rhizosphere which consequently impairs its growth (Burdick, 1989;Bertness, 1991;Curado et al., 2020), and naturally limits this species distribution to the high tide border of salt marshes. Thus, in the present work, we tested the application of a physical barrier that would prolong waterlogging around the plants and evaluated the impact on its physiology.

Plant Material and Collection Site
Sampling occurred before the start of the growing season (February 2020) in the Hortas salt marsh (Alcochete;38 • 45.661' N,8 • 56.116' W), located in the middle estuary, adjacent to the Tagus Estuary Natural Reserve (Figure 1). This salt marsh is flooded twice per day, being dominated by the halophyte species Spartina maritima in the lower marsh (circa 12% coverage), Halimione portulacoides in the mid-upper marsh (circa 35% coverage), and Sarcocornia fruticosa in the upper marsh (circa 20% coverage) . More recently, the halophyte invasive species S. patens has managed to establish itself in the margins, justifying the importance of the current study.

Applied Treatments
Spartina patens plants were subjected to two treatments: (1) control, where the plants did not undergo any intervention, and (2) the waterlogging treatment (waterlogged) where plant turfs filled the inside of a 50 cm long and 10 cm diameter PVC tube buried down into the sediments, leaving an 8 cm margin outside, which allowed the tidal water to enter the tube and remain longer inside and, thus, increase the waterlogging of the plants (n = 5 for each treatment). The experiment lasted for 30 days. Several field measurements were made, namely, in vivo chlorophyll a pulse amplitude modulated (PAM) fluorometry measurements. Moreover, 20 leaves from each treatment were also collected directly into individual tubes with liquid nitrogen (see sections below). Finally, whole plants of S. patens from both treatments were also collected, transported to the lab, washed, removed from excess water, weighted (fresh weight), and oven-dried at 60 • C to constant weight (dry weight).

Chlorophyll a Pulse Amplitude Modulated Fluorometry
Ten leaves from plants from each treatment were dark-adapted for 15 min and PAM measurements were performed using a FluoroPen FP100 (Photo System Instruments, Czechia). For the analysis of chlorophyll transient light curves (Kautsky plot), and the derived fluorometric parameters (Table 1), the JIP-test was used .

Leaf Infrared Thermography
Thermal images were obtained with a FLIR E50bx infrared camera (FLIR Systems, Inc., Wilsonville, OR, United States) producing images of 320 × 240 resolution with an accuracy of ±0.045 • C. Ten leaves were randomly selected from plants of both treatments, having a water bottle at ambient temperature near the leaves as reference. The average temperature of each leaf was calculated on each image. All image processing and analysis were performed in FLIR Tools software (version 6.4.18039.1003, FLIR Systems, Inc.).

Root Respiration
Fresh fine roots of each treatment weighing approximately 100 mg were transferred to 10-mL reaction tubes, and 6 mL of TTC-solution [0.6% (w/v) triphenyl-tetrazolium chloride in 0.06 M Na 2 HPO 4 -KH 2 PO 4 and 0.05% (v/v) Tween 20] was added to each tube (n = 5) (Brunner et al., 2002). In duplicate tubes, 0.15 mM KCN was added to determine the inhibited respiration. The samples were then incubated for 24 h at 25 • C. After incubation, the TTC solution was decanted and triphenyl formazan (TPF) extraction was made by adding 2 mL of ethanol and boiling at 80 • C for 15 min (Ruf and Brunner, 2003). After collecting the supernatant in new tubes, the absorbance of 1 mL was measured at 520 nm with a spectrophotometer (UV500 UV-Visible Spectrometer, Unicam, Waltham, MA, United States). The root residues in the test tubes were dried at 80 • C for 72 h and weighed. Reduction of TTC was calculated as µg of TPF produced per hour per g dry weight (DW).

Proline Quantification
Proline content was determined according to Bates et al. (1973). For each treatment, plant leaves (n = 5) were homogenized in 3% aqueous sulfosalicylic acid and the homogenate centrifuged a 9,000 g for 15 min at 0 • C (Sigma 2-16K, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany). The supernatant was collected, and the reaction consisted of 2 mL of extract combined with 2 mL of glacial acetic acid and 2 mL of acid ninhydrin. The reaction occurred for 1 h at 100 • C, after which the reaction was stopped in an ice bath. The reaction mixture was extracted with 4 mL of toluene and its absorbance read at 520 nm with a spectrophotometer (UV500 UV-Visible Spectrometer, Unicam, Waltham, MA, United States) and compared with a standard curve of proline, expressed in µmol g −1 fresh weight (FW).

C4-Photosynthetic Carboxylating Enzymes Activity and Pigment Analysis
Carboxylating enzymes, PEPC and Rubisco, were extracted from frozen leaf samples according to Carmo-Silva et al. (2008), except that 50 mM HEPES-KOH pH 7.3 was used and 0.5% (v/v) Triton X-100 added. Briefly, approximately 50 mg FW were extracted in a cold mortar containing quartz sand, 1% (w/v) insoluble polyvinylpyrrolidone (PVP) and 1 mL of ice-cold extraction medium [50 mM HEPES-KOH pH 7.3, 1 mM EDTA, 5% (w/v) PVP25000, 6% (w/v) polyethylene glycol (PEG4000), 10 mM dithiothreitol (DTT), 1% (v/v) protease cocktail inhibitor (Sigma, St Louis, MO, United States) and 0.5% (v/v) Triton X-100]. After taking aliquots for pigment analysis, the homogenate was centrifuged for 3 min at 16,800 g at 4 • C (Sigma 2-16K, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany) and the supernatant (crude extract) was kept at 4 • C and immediately used for measuring the activities of Rubisco (EC 4.1.1.39) and PEPC (EC 4.1.1.31). The activities of Rubisco were assayed at 25 • C by 14 CO 2 incorporation into acid-stable products according to Parry et al. (1997) to with modifications (Correia et al., 2020). The assay medium (1 mL per sample) contained 50 mM Bicine-KOH pH 8.2, 40 mM MgCl 2 , 10 mM NaH 14 CO 3 (7.4 kBq µM −1 ) and 0.4 mM ribulose-1,5-bisphosphate (RuBP). To measure Rubisco initial activity (V i ), 25 µL of crude extract was added to assay medium, and the reaction stopped after 1 min with the addition of 100 µL of 1 M HCl. To determine Rubisco total activity (V t ), 25 µL of crude extract was added to assay medium without RuBP for 3 min, to allow the carbamylation of enzyme catalytic sites. Rubisco V t reaction was started by adding RuBP and stopped after 1 min with the addition of 100 µL of 1 M HCl. The mixture was completely dried at 60 • C after which the residue was resuspended in 0.5 mL of distilled water and mixed with 5 mL of scintillation liquid (BioSafe LS Cocktail, Beckman, United States). Radioactivity of the 14 C incorporated in the acidstable products was measured by scintillation counting (LS 7800 spectrophotometer, Beckmann Instruments Inc., Fullerton, CA, United States). Rubisco activation state (%) was determined by the V i /V t ratio.
PEPC physiological (V physiol ) and maximum (V max ) activities were measured in a continuous assay at 340 nm and 25 • C (UV500 UV-Visible spectrophotometer, Unicam, Cambridge, United Kingdom) according to Bakrim et al. (1992) with some modifications (Carmo-Silva et al., 2007). The reaction mixture for V physiol (1 mL) consisted of 50 mM HEPES-KOH pH 7.2, 10 mM MgCl 2 , 10 mM NaHCO 3 , 2.5 mM PEP (Sigma), 12 units of MDH (Sigma) and 20 µL of crude extract. For V max , the reaction mixture consisted of 50 mM HEPES-KOH pH 8.0, 10 mM MgCl 2 , 10 mM NaHCO 3 , 10 mM PEP (Sigma), 12 units of MDH (Sigma) and 20 µL of extract. In both cases, the reaction was started by the addition of 0.2 mM (final concentration) NADH (Sigma). Each measured activity is the mean of three replicate on the same extract. PEPC activation state (%) was calculated as V physiol /V max ratio.
For pigment analysis, each 20 µL aliquot previously retrieved from the leaf extract was diluted in 980 µL of methanol. After mixing in the vortex, the samples were left in the dark at 4 • C overnight. After centrifuging for 1 min at 13,000 g at 4 • C (Sigma 2-16K, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany), the absorbance at 470, 652.4, 665.2, and 700 nm were measured in an Epoch TM 2 Microplate Spectrophotometer (BioTek, Winooski, VT, United States). Pigment concentrations were determined according to the equations in Lichtenthaler and Buschmann (2001).

Antioxidant Enzyme Assays
To extract the soluble protein fraction, leaf samples were grinded in a cooled mortar with 0.5 mL of 50 mM sodium/potassium phosphate extraction buffer (with 0.1 mM Na-EDTA, 2 mM PVP, 10 mM DTT, 0.1 mM PMSF, and 24 µM NADP, pH 7.6).
The homogenate was centrifuged at 13,000 g for 10 min at 4 • C (Sigma 2-16K, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany) and the supernatant was collected to a new tube. Protein concentration was determined according to Bradford (1976) in an Epoch TM 2 Microplate Spectrophotometer (BioTek, Winooski, VT, United States).
Catalase (CAT; EC 1.11.1.6) activity was measured according to Teranishi et al. (1974), through H 2 O 2 consumption monitoring and the decrease in absorbance at 240 nm (ε = 39.4 mM −1 cm −1 ). The reaction mixture contained 50 mM of sodium/potassium phosphate buffer (pH 7.0), and 30 mM of H 2 O 2 with the reaction being started by the addition of 5 µL of extract. Ascorbate peroxidase (APx; EC 1.11.1.11) was assayed according to Tiryakioglu et al. (2006). The reaction mixture contained 50 mM of sodium/potassium phosphate buffer (pH 7.0), 0.1 mM of H 2 O 2 , and 0.25 mM L-ascorbate, and the reaction was also initiated with the addition of 5 µL of the extract. The activity was recorded as the decrease in absorbance at 290 nm and the amount of ascorbate oxidized calculated from the molar extinction coefficient (ε = 2.8 mM −1 cm −1 ). Guaiacol peroxidase (GOPx; EC 1.11.1.7) activity was assayed according to Mika and Lüthje (2003) through the monitorization of guaiacol oxidation at 470 nm (ε = 26.6 mM −1 cm −1 ). The reaction mixture contained 50 mM of sodium/potassium phosphate buffer (pH 7.0), 10 mM of H 2 O 2 , and 8 mM guaiacol, and the reaction was initiated with the addition of 5 µL of the extract. Superoxide dismutase (SOD; EC 1.15.1.1) activity was assayed according to Marklund and Marklund (1974) by measuring the reduction of pyrogallol at 325 nm. The reaction mixture contained 30 mM of sodium/potassium phosphate buffer (pH 7.0) and 0.24 mM of pyrogallol, with the reaction being started by the addition of 5 µL of extract. Glutathione reductase (GR; EC 1.8.1.7) activity was assayed according to Edwards et al. (1990) by measuring the fall in absorbance at 340 nm as NADPH was oxidized (ε = 6.22 mM −1 cm −1 ). The reaction mixture contained 25 mM of sodium/potassium phosphate extraction buffer (pH 7.6), 0.5 mM of oxidized glutathione and 0.2 mM NADPH, starting the reaction by the addition of 5 µL of extract. Control assays were done in the absence of substrate to evaluate the autoxidation of the substrates. All assays were performed in a total volume of 200 µL per well at 25 • C in an Epoch TM 2 Microplate Spectrophotometer (BioTek, Winooski, VT, United States).

Lipid Peroxidation Analysis
Lipid peroxidation products were determined as previously described (Heath and Packer, 1968). Leaves were homogenized briefly in 1.5 mL of 10% (v/v) trichloroacetic acid (TCA), containing 0.4% (w/v) thiobarbituric acid (TBA). The reaction was conducted at 100 • C for 30 min, being halted through placement in ice. After centrifugation at 15,000 g for 10 min at 4 • C (Sigma 2-16K, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany), 1 mL of the supernatant was collected and mixed with 1 mL of 0.4% TBA and incubated again under the same conditions. The absorbance of the supernatant was recorded at 532 and 600 nm by spectrophotometry (UV500 UV-Visible Spectrometer, Unicam, Waltham, MA, United States).

Fatty Acid Profiles
The analysis of fatty acid was performed by direct transesterification of leaf samples, in freshly prepared methanol sulphuric acid (97.5:2.5, v/v), at 70 • C for 60 min, using the internal standard pentadecanoic acid (C15:0) (Feijão et al., 2018). Fatty acid methyl esters (FAME) were recovered using petroleum ether, dried with an N 2 flow, and re-suspended in an adequate amount of hexane. The FAME solution was analyzed through gas chromatography (Varian 430-GC gas chromatograph equipped with a hydrogen flame ionization detector set at 300 • C, Middelburg, Netherlands), by addition of 1 µL, setting the injector temperature to 270 • C, with a split ratio of 50. The fused-silica capillary column (50 m × 0.25 mm; WCOT Fused Silica, CP-Sil 88 for FAME; Varian, Middelburg, Netherlands) was maintained at a constant nitrogen flow of 2.0 mL min −1 and the oven set to 190 • C. Fatty acids identification was performed by comparison of retention times with standards (Sigma-Aldrich) and chromatograms were analyzed by the peak surface method, using the Galaxy software. The double bond index (DBI) was calculated, to determine the membrane saturation levels, as previously described (Feijão et al., 2018):

Statistical Analysis
Since the data lacked normality and homogeneity, the statistical analysis was based on Mann-Whitney non-parametric tests (GraphPad Prism 8.4.2 for Windows, GraphPad Software, San Diego, CA, United States). Multivariate statistical analyses [SIMPER and Canonical Analysis of Principal Coordinates (CAP)] were performed using Primer 6 software (Clarke and Gorley, 2006). The data obtained from the Kautsky plots, thermography data, oxidative stress and fatty acids were used  as the basis for the construction of the respective resemblance matrixes based on the Euclidean distances between samples. To evaluate the different metabolic datasets obtained as a whole (in opposition to univariate analysis), statistical multivariate models based on Kautsky plots, thermography data, oxidative stress and fatty acids were generated using Principal Coordinates Analysis (PCO) (Clarke and Gorley, 2006).

Chlorophyll a PAM Analysis
Observing the Kautsky plots resultant from the in vivo PAM fluorometric analysis, lower fluorescence values could be observed in plants subjected to the treatment when compared with control plants (Figure 2).   waterlogged plants, although that decrease was only statistically significant for TR 0 /CS and ET 0 /CS. It was also a similar reduction in the number of oxidized PS II reaction centers (RC/CS) (Figure 3E).
Further analysis of the functioning of different components of the photosystems and ETC (Figure 4) showed a decrease in the oxidized quinone pool size in the waterlogged plants, followed by an enhancement in the number of QA redox turnovers until maximum fluorescence was reached (N). Although no significant changes were observed in the energy needed to close all RCs (S M ), there was a decrease in the probability of a PS II chlorophyll molecule functioning as a RC (γ RC ) in the treated plants. However, no significant differences were observed in the QA reduction rate (M 0 ).
Although the active oxygen-evolving complexes (OEC) showed no differences between control and waterlogged plants (Figure 5A), the P G , the grouping probability that correlates with the disconnection between the two PS II units, increased in the later plant group (Figure 5B).
Regarding PS II and PS I, waterlogged plants presented a significant decrease in photochemical processes, both in the contribution of light (TR 0 /DI 0 ; Figure 5H) and dark [ψ 0 /(1 -ψ 0 ); Figure 5C] reactions of the photochemical cycle. On the other hand, at the PS I level there was a significant increase in the activity of this photosystem [δ R0 /(1 -δ R0 ); Figure 5F] in the intervened plants, although there was a decrease in the equilibrium constant for the redox reaction between both photosystems toward the PS II [ψ E0 /(1 -ψ E0 ); Figure 5D]. Nevertheless, there were no significant changes in the reaction center density within the PS II antenna chlorophyll bed (RC/ABS; Figure 5G) or in the electron transport from PQH 2 to the reduction of the PS I end acceptors (RE 0 /RC; Figure 5E).
In summary and observing the structure functional indexes, there was a decrease of the photochemical processes ( Figure 6A) and an increase of the non-photochemical or dissipative processes (Figure 6B) in the waterlogged plants.

Leaf Thermography
Regarding leaf surface temperature measured through infrared thermography, there was a statistically significant increase in temperature in the treated plants (11.06 • C) relatively to control ones (10.77 • C) (Figure 7).

Root Respiration
An increase of root respiration in waterlogged plants was measured (Figure 8A), resulting not from changes in the alternative oxidase (AOX) activity ( Figure 8B) but rather from a twofold increase in KCN-sensitive respiration ( Figure 8C).

Proline
Leaves of waterlogged plants of S. patens experienced a fivefold increase in proline (Figure 9), to circa 20 µmol g −1 FW, contrasting with the leaves of control plants with circa 4 µmol g −1 .

C4-Photosynthetic Carboxylating Enzymes and Pigment Analysis
Relatively to the C4-photosynthetic carboxylating enzymes activities, no statistically significant changes were observed either in the activation state, or maximal (V max ) and physiological (V physiol ) rates of PEPC (Supplementary Figure 1), or the activation state, or initial (V i ) and total (V t ) rates of Rubisco (Supplementary Figure 2). The same was observed for pigments Chl a, Chl b, total chlorophyll, and carotenoids (Supplementary Figure 3).

Antioxidant Enzyme Analysis
Regarding the antioxidant enzymes CAT, APx, GOPx, SOD, and GR (Figure 10), only the latter two showed a significant decrease in activity on waterlogged plants, with a decrease of twofold in SOD and almost a threefold in GR.

Lipid Peroxidation and Fatty Acid Profile
While lipid peroxidation products increased in waterlogged plants ( Figure 11A), no changes occurred in total fatty acids content ( Figure 11B).
Regarding the relative abundance of fatty acids, there was a decrease in triunsaturated hexadecatrienoic acid (16:3), exclusively present in plastidial lipids, whilst there was an increase in stearic acid (C18:0) of waterlogged plants relative to controls (Figure 12). Although there were no significant changes in the major fatty acids classes (Supplementary Figure 4), the ratio of saturated to unsaturated fatty acids ratio (SFA/UFA) increased in waterlogged plants (Supplementary Figure 5).

Biophysical and Biochemical Profiles
The multivariate PCO analysis allowed to distinguish the control from the waterlogged plants of S. patens using the Kautsky plot, leaf thermography, oxidative stress data and fatty acid profiles (Figures 13, 14), which explain the variables' total variation with various degrees. Through the simple application of remote tools (Figure 13), the Kautsky plot first PCO axis explains 98.9% of the whole variation between the two evaluated groups (Figure 13A), while due to a higher dispersion along the second PCO axis, the thermography data PCO1 has a lower explanatory power (64.8% of total variation) ( Figure 13B).
Observing the biochemical profiles, the oxidative stress measurements reveal good separation of the two groups (87.3% of total variation), along only one axis of the PCO (Figure 14A). On the other hand, when evaluating the samples fatty acid profiles, it was possible to observe a dispersion along the two PCO axis, indicative of a higher dispersion in the biochemical characteristics of the samples and lower explanatory power (46.2% of total variation) ( Figure 14B).

DISCUSSION
The impacts of a highly aggressive and invasive species, such as S. patens urges the need to develop measures, to counteract its spreading. Control measures should be directed to this species and are less harmful to the surrounding environment and co-occurring species. Our proposed alternative method for eliminating the invasive halophyte S. patens proved to be a very effective method to significantly increase the stress levels of the plants in 30 days, using the species physiology drawbacks. Results provide new insights for an ecoengineering solution to be applied in future S. patens control measures in intertidal systems. The fact that S. patens do not tolerate waterlogging is the only limiting factor that is currently preventing full occupation of the intertidal border (Burdick, 1989;Bertness, 1991;Curado et al., 2020), and fully replacing other halophyte native species. Furthermore, once the NIS is removed, indigenous species, such as H. portulacoides and S. fruticosa can reoccupy that area and restore habitat functionality (Curado et al., 2020). Thus, any tool that facilitates this NIS dieback, is of added value to recover the ecosystem floristic diversity.
The effect of waterlogging on S. patens was very clear in different photochemical-related variables and showed lower photosynthetic efficiency when compared with the nonintervened plants, although no significant changes were observed in C4 photosynthesis-related enzymes PEPC and Rubisco as well as in the pigment profile.
The observation of the two distinct treatments of S. patens showed a depression in the Kautsky curves of the waterlogged plants, presenting lower fluorescence values, a characteristic of stress effects (Duarte et al., 2015). Energy fluxes suffered a major impact at the level of electron transport (ET 0 /CS), with a reduction in the quinone pool size, given by the area above the transient curve (Strasser et al., 1995;Joliot and Joliot, 2002), and a decrease in the number of available reaction centers (RC/CS). This leads to an excessive energy accumulation (Kalaji et al., 2011) that can destroy the D1 protein, and eventually impair all photochemical machinery (Demmig-Adams and Adams, 1992;Qiu et al., 2003;Duarte et al., 2013). To prevent oxidative damage in the chloroplasts, this high energy accumulation needs to be dissipated, frequently through the xanthophyll cycle (Demmig-Adams and Adams, 1992). However, no changes were observed in the carotenoid pool, occurring a decrease in photochemical  quenching and an increase in non-photochemical quenching (NPQ), leading to a decrease in the lumen pH, attributed to protonation of the light-harvesting complexes (LHC) proteins associated with the PSII (Cousins et al., 2002). This, in turn, can explain the increase in leaf temperature observed in the waterlogged plants (Zhang et al., 2019). Moreover, it was possible to observe that waterlogged plants had their trapped and transported energy fluxes reduced, but no increase in the energy dissipation. Under stress, a common feature observed is a maintenance of the absorbed and trapped energy flux, but a low energy use in the electron transport, leading to energy bursts inside the chloroplast that are normally diverted to dissipation in the form of heat or fluorescence (Duarte et al., 2016). In this case, only a reduction of the trapped and energy transport fluxes indicating a lower ability of the plants to harvest light, a condition that can promote leaf senescence and plant mortality.
With the roots under water for a more prolonged time, it would be expected to observe a decrease in root respiration (Pan et al., 2021). However, S. patens can develop aerenchyma, pumping atmospheric oxygen through the leaves to ventilate the roots and, thus, allowing it to increase its respiration (Burdick and Mendelssohn, 1987;Burdick, 1989). Nevertheless, and considering the lower energy harvested by waterlogged plants, increased root respiration and consequent consumption of energy reserves (such as carbohydrates), will inevitably increase the energetic stress of the plants. Spartina patens also presented an increased proline content in its leaves, increasing leaf water potential to counteract the differential osmotic pressures between soil and leaves (Briens and Larher, 1982;Koyro et al., 2006;Aziz et al., 2011). This mechanism also poses significant energetic costs to the individuals, that while trying to counteract the waterlogging osmotic constraints, end up increasing the degree of energetic stress of the plants (Iqbal et al., 2014).
Oxygen deprivation will lead inevitably to the increase of intracellular reactive oxygen species (ROS) due to metabolism impairment under stress (Bailey-Serres and Chang, 2005; FIGURE 12 | Fatty acid profile in leaves of control plants (white boxes) and waterlogged plants (gray boxes) of Spartina patens after 30 days (n = 5, different letters indicate significant differences at p < 0.05). Pucciariello et al., 2012), severely damaging, amongst others, proteins, and lipids of cells membranes (Sharma et al., 2012;Baxter et al., 2014). Nevertheless, under waterlogging no increase in the antioxidant enzymatic activity could be observed, thus contributing to the accumulation of ROS inside the cells of the waterlogged plants. In fact, two first-line defense enzymes (SOD and GR) had their activities severely reduced. All these impairments of the antioxidant enzymatic system are concomitant with the observed increase in lipid peroxidation. Once again, this will impose severe energetic stress, due to damage in the membranes of the energetic organelles and thus impacting the electronic transport systems and the cell energy budgets. Additionally, exposure to waterlogging also induced changes in the fatty acid profile of S. patens, which increased the ratio between saturated and unsaturated fatty acids, mainly contributed by an increase in stearic acid (C18:0), which likely resulted in a decrease in membrane fluidity. Moreover, there was a significant decrease in the triunsaturated hexadecatrienoic acid (16:3), which is present in the major plastidial lipid monogalactosyldiacylglycerol and can be linked to lower photosynthetic efficiency measured in waterlogged plants (Kobayashi et al., 2016).
Using multivariate analysis of the assessed biophysical and biochemical traits it was possible to observe that techniques, such as PAM and infrared thermography, are the most efficient tools to evaluate in situ the success of the intervention and the stress levels imposed on the plants targeted to be removed. As in previous ecological studies in response to stress (Cruz de Carvalho et al., 2020a,b;Duarte et al., 2020), the use of bio-optical techniques (e.g., Kautsky plot) allowed an efficient identification of the waterlogged plants and, thus, provide efficient and reliable tools to monitor the success of control interventions, without disturbing the process itself and allowing repeated measures over the intervention period.
In terms, of the ecoengineering solution tested in the present work, the physiological changes observed indicate that the adopted strategy can be an eco-friendly control measure, increasing the plants stress levels significantly, although in the considered 30-day period there was no plant mortality, indicating that these control measures should be performed in more prolonged periods.

CONCLUSION
The application of a simple barrier to allow waterlogging proved to be an effective and low-cost solution to impair the physiology and biochemical pathways of the invasive halophyte S. patens. Although long-term studies need to be performed, this ecoengineering solution has the potential to control and eliminate this species from salt marshes and other intertidal systems in future ecosystem restoration programs. Since S. patens turfs are easily identified, the upscaling of this technique could involve the application of tubes with different diameters according to plant turfs density, allowing waterlogging to be prolonged in space and time. Furthermore, the application of a simple bio-optical tool will allow the stakeholders to easily follow the process of suppressing the species without the interference of the process and allowing repeated measures over the intervention period.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
BD, VF, and PR-S: conceptualization. BD: methodology, supervision, project administration, and funding acquisition. RC and BD: formal analysis. RC, EF, ID, VP, MS, AS, and AM: investigation. RC: data curation and writing-original draft preparation. EF, ID, VP, MS, AM, AS, IC, PR-S, VF, and BD: writing-review and editing. All authors have read and agreed to the published version of the manuscript.