Potential Distribution of Seagrass Meadows Based on the MaxEnt Model in Chinese Coastal Waters

WANG Ming, WANG Yong, LIU Guangliang, CHEN Yuhu, and YU Naijing

Potential Distribution of Seagrass Meadows Based on the MaxEnt Model in Chinese Coastal Waters

WANG Ming1), *, WANG Yong1), LIU Guangliang2), CHEN Yuhu1), and YU Naijing2)

1),,266200,2),,250101,

Seagrass meadows are generally diverse in China and have become important ecosystem with essential functions. How- ever, the seagrass distribution across the seawaters of China has not been evaluated, and the magnitude and direction of changes in seagrass meadows remain unclear. This study aimed to provide a nationwide seagrass distribution map and explore the dynamic changes in seagrass population under global climate change. Simulation studies were performed using the modeling software MaxEnt with 58961 occurrence records and 27 marine environmental variables to predict the potential distribution of seagrasses and calculate the area. Seven environmental variables were excluded from the modeling processes based on a correlation analysis to ensure predicted suit- ability. The predicted area was 790.09km2, which is much larger than the known seagrass distribution in China (87.65km2). By 2100, the suitable habitat of seagrass will shift northwest and increase to 923.62km2. Models of the sum of the individual family under-pre- dicted the national distribution of seagrasses and consistently showed a downward trend in the future. Out of all environmental vari- ables, physical parameters (., depth, land distance, and sea surface temperature) contributed the most in predicting seagrass distri- butions, and nutrients (., nitrate, phosphate) ranked among the key influential predictors for habitat suitability in our study area. This study is the first effort to fill a gap in understanding the distribution of seagrasses in China. Further studies using modeling and biolo- gical/ecological approaches are warranted.

seagrass meadows; species distribution modeling; global climate change; Chinese coastal waters

1 Introduction

Seagrass meadows are valuable coastal and marine eco- systems on the planet, providing a range of critical environ- mental, economic, and social benefits (Cullen-Unsworth and Unsworth, 2018; Jayathilake and Costello, 2018). While covering approximately 0.1% of the Earth’s seafloor, sea- grass meadows support a wide range of biodiversities, sta- bilize sediment, filter water, provide coastal protection, pro-duce more oxygen than rainforests, form the basis of theworld’s primary fishing grounds, and play a vital role in mitigating climate change and stabilizing the carbon cycle (Duarte., 2013; Unsworth., 2018). However, sea-grass has been disappearing globally at a rate of 7% peryearsince 1990. Now totally more than 29% of seagrassbeds have been lost, and approximately 14% of the speciesare at risk of extinction (Waycott., 2009; Duarte., 2013). In addition, seagrasses have received comparative- ly little consideration in scientific research and conservation, and seagrass meadows are among the world’s least-known ecosystems (Unsworth., 2018).

Globally, the cumulative impacts of multiple stressors, such as coastal development, population growth, serious pollution, and climate change, play an important role in de- termining the distribution of the seagrass meadows at pre- sent and in the future (Adams., 2020). Several pre- vious studies focused on seagrass meadow monitoring (Short., 2014; Telesca., 2015), genetic evolution (Ol- sen., 2016; Kendrick., 2017), species and dis- tribution management, and restoration practices (Kenwor- thy., 2018), have provided scientific basics for the protection of seagrass population. However, increasing re- cognition of the conservation importance of seagrasses is calling for accurate estimates of global seagrass distribution. Thus, scientists and ocean and coastal managers have focus- ed on mapping the distribution of seagrasses (Short., 2011) by using different methods, such as remote sensing (Gumusay., 2019), aerial photography, and underwa- ter videography (Schultz, 2008). With the increasing model development and data accumulation in recent years, com- puterized modeling has become a significant research me- thod for seagrass population management and conserva- tion (Jayathilake and Costello, 2018; Staehr., 2019). Jayathilake and Costello (2018) modeled the global distri- bution of seagrass biome by combining over 43000 occur- rence records and 13 environmental variables in the Max- Ent modeling software with 30 arcsec resolution and up-dated its predicted distribution area to 1646788km2. By ra-tionalizing and updating a range of existing datasets of sea- grass distribution around the globe, McKenzie. (2020)estimated with moderate-to-high confidence that the global seagrass area is 160387km2.

Obviously, the distribution of seagrasses is influenced by different parameters, such as physical variables (., wa- ter depth, temperature, salinity, water clarity, wave height, and photosynthetically active radiation), chemical parame- ters (., pH, phosphate, nitrate, and dissolved oxygen con- centration), and biological factors (competition, predation, and genetic factors) (Unsworth., 2019). Different sea- grass species usually have relatively specific distribution areas; for instance, the eelgrassL. is wide-ly distributed in shallow waters of the northern hemisphere (Olsen., 2016), whereasis an en- demic seagrass to the Mediterranean Sea (Houngnandan., 2020). Given the successful application of species dis-tribution models (SDMs) in the marine realm (Martínez., 2018), the variability of the offshore environment and the specificity of seagrass distribution make the model re- quirements more complicated (Assis., 2018). In par- ticular, climate change is the most widespread threat to sea- grass ecosystems, and the distribution pattern of seagrass and its relationship with marine environment changes are key problems that need to be solved (Chefaoui., 2018; Olsen., 2018; Unsworth., 2019).

Compared with studies in Australia, Europe, and North America, studies on seagrass in China are still at the start- ing stage (Larkum., 2018). In addition, the lack of ba- sic information hinders national conservation and restora- tion programs for seagrasses (Zheng., 2013). Several researchers suggested that four families, 10 genera, and 22species of seagrasses are distributed in Chinese coastal wa- ters, with a predicted area of 87.65km2(Zheng., 2013;Meng., 2019). In recent years, the researchers fromChina have participated in some international seagrass mo- nitoring programs (., SeagrassNet) and investigated sea-grass population dynamics (Zhang., 2019), physio-logical ecology and evolutionary mechanism (Yu., 2018),and protection and restoration (Wang., 2019; Jiang., 2020). In 2015, a National Science and Technology Basic Work Program (2015FY110600) was funded for a na- tional seagrass habitat survey in China, and many new re- corded seagrass populations were found in different sea areas (Zhang., 2019). However, an evaluation of seagrass distribution across China has not been conducted, and the magnitude and direction of changes in seagrass remain un- clear.

The present study aimed to provide a nationwide sea- grass distribution map and to explore the dynamic changes of the seagrass population under the condition of climate change in China. We applied an SDM approach that com- bines information from different marine environmental con-ditions and seagrass occurrence data. This study represents the first attempt to map the environmental suitability for seagrasses in China under current conditions and future cli-mate scenarios, which allows a fine-scale definition of prio- rity areas in seagrass conservation measures for now and in the future.

2 Methods

2.1 Species Occurrence Data

As mentioned in the Introduction, 22 species of seagrass-es from the families Cymodoceaceae, Hydrocharitaceae, Ruppiaceae, and Zosteraceae are distributed in Chinese coas- tal waters. Hence, this study focused on species belonging to these families (Table 1).

Table 1 List of seagrass species used in this study, and the number of seagrass occurrence data is given in brackets

Seagrass occurrence data were from the Global Biodi- versity Information Facility (GBIF, 2017) and Ocean Bio- geographic Information System (Wang., 2021). Taxo- nomic names were reconciled with the World Register of Marine Species (Horton., 2020) and the Atlas of Sea- grass (Green and Short, 2003). In case the insufficient oc- currence records in China will reduce the simulation accu-racy, the global seagrass occurrence data were downloaded for this study. In addition, we focused the current analysis during 2011–2020 to match the time scale of marine en- vironmental datasets. Occurrence data were accessed as spe- cies-specific tables of latitude/longitude and imported into ArcGIS version 10.4 (ESRI, Redlands, California) with a WGS84 coordinate system (Bittner., 2020). After datapreprocessing for excluding records with coordinate un- certainty, records falling on land, and duplicate records, the remaining 58961 species occurrence data were used in the subsequent modeling (Fig.1).

Fig.1 Global seagrass occurrence data used in this study.

2.2 Marine Environmental Data

Twenty-seven abiotic variables related to the distribution of seagrasses were chosen as modeling parameters from the Global Marine Environment Datasets (GMED), which are the publicly available climatic, biological, and geophysicalenvironmental layers featuring the present, past, and future environmental conditions (Basher., 2018; Table 2). Allthe layers were annual averages calculated over decadeswith a five arc-min spatial resolution (approximately 9.2 km at the equator) initially (Basher., 2018; Table 2). All interpolated layers were delimited to 100m-depth thres- holds for seagrass occurrence along Chinese coastal wa- ters based on GEBCO_2019 Grid (http://www.gebco.net/) and then re-interpolated to 30 arc seconds (Jayathilake and Costello, 2018) to ensure model accuracy.

The past(last glacial maxima, 22Myr) and future(the year 2100) abiotic layers were also from GMED, both of which contained four parameters (Basher., 2018) and were preprocessed as described above.

Table 2 List of environmental variables used in this study from GMED

Notes:†The selected variables in the final model were marked by ‘√’; ‘–’, no unit.

2.3 Modeling

The MaxEnt species distribution modeling software (ver- sion 3.4.1) was used to generate the seagrass SDM (Phil- lips., 2006, 2017). During data exploration, explana- tory variables were assessed for correlation. When two va- riables were correlated at Pearson’s>0.7, the variable which was the most proximal in determining the distribu- tion of seagrasses was selected (Ferrari., 2018; Table 3). To obtain alternate estimates of which variables were the most important for different families and all seagrass species, we performed a jackknife test to improve predic- tive performance. In general, variables with a permutation importance of zero were deleted, and the remaining ones were more likely to be directly relevant to the species be- ing modeled (Phillips., 2017).

Table 3 Correlation analysis of different variables

In the current study, MaxEnt models were generated us- ing 10 cross-validated replicate runs with the following pa- rameters: random test percentage=25, regularization mul- tiplier=1, max number of background points=10000, ma- ximum iterations=1000, and convergence threshold=10−5. MaxEnt models for each family and all seagrass species were generated separately to identify distribution differ-ences between families. Post image processing for the map was carried out using ArcMap version 10.4, and areas with a probability value>0.5 were taken as the potential distri- bution of seagrasses (Jayathilake and Costello, 2018).

Projections for the past (last glacial maxima, 22Myr) and future (the year 2100) were applied to make reliable predictions of distribution under climate change. The po- tential distribution maps were cropped to Chinese coastal waters, and the geographic extent was 18˚–42˚N, 108˚–124˚E. Finally, centroids of the past, present and future pro- jections were determined. The distance and direction be- tween centroids were used to estimate how the distribution range will shift in respect to the present one. The receiver operating characteristic and area under the receiver ope- rating characteristic curve (AUC) were assessed to test the performance of the model (Phillips., 2006).

3 Results

3.1 Environmental Variables

The relative importance of the environmental variables showed obvious divergences between each family and all seagrasses (shown in Fig.2). For Cymodoceaceae, water depth (42.4%), minimum sea surface temperature (19.5%), and nitrate (17.2%) had the highest contribution in sea- grass distribution modelings. For Hydrocharitaceae, the most important variables were water depth (34.1%), cal- cite (15.8%), and dissolved oxygen (13.7%). Nevertheless, mean sea surface temperature (31.4%), land distance (16.4%), and phosphate (14.5%) were much more important for Rup- piaceae distribution modeling. For Zosteraceae and all sea- grasses, land distance (39.3%), minimum sea surface tem- perature (25.4%), maximum sea surface temperature (18.3%), water depth (46.7%), mean sea surface temperature (22.3%), and land distance (19.1%) were the three most critical va- riables, respectively (Fig.2). Furthermore, variables with a permutation importance of 0 were deleted to perform specific distribution modeling for each family. Hence, the MaxEnt model showed that the most effective single va- riable of predicting the distribution of seagrasses had ob- vious inter-family differences in Chinese coastal waters.

Fig.2 Spider plots of the permutation importance (%) of each environmental variable within species distribution models.

3.2 Predicted Distribution and Areas

The MaxEnt model for each family and all seagrasses demonstrated high predictive strength, indicated by AUC =0.961±0.036 (mean±standard deviation; Table 4).

The MaxEnt model for family Cymodoceaceae indicated a probability of distribution from 0.500 to 0.584, with a pre- dicted area of 20.54km2along the southern Hainan and western Taiwan coast (Table 4 and Fig.3A). In 2100, the dis- tribution areas of Cymodoceaceae will decrease slightly to 16.26km2(Table 4 and Fig.3C), and no suitable area in China was found during the last glacial maximum (Table 4 and Fig.3B).

The family Hydrocharitaceae showed good agreement with Cymodoceaceae but had a larger and relatively stable distribution area (about 33.38km2) in China at present and in the future (Table 4, Figs.3D and 3F).

The family Ruppiaceae was mainly distributed in the coastal areas along the East China Sea, and the predictedarea was 74.47km2at present, which might decrease by about 10% in 2100 (Table 4, Figs.3G and 3I).

The family Zosteraceae was predicted to have the largest potential distribution in China coastal waters. It was more than the sum of the three other families (412.59km2; Ta- ble 4), and mainly distributed in the Bohai Sea, the Yel- low Sea, and the East China Sea (Fig.3J). The predicted distribution showed a trend of migration to the north by 2100, and the area would reduce to 372.36km2(Fig.3L). In addition, Zosteraceae was the only family predicted to be distributed in China during the last glacial maximum, with the area reaching 10.27km2(Table 4 and Fig.3K).

When considering all the seagrass species, the predict- ed areas would increase to 790.09km2and reach 923.62 km2by 2100 (Table 4, Figs.3M and 3O).

Notably, the maps derived from the four families show- ed an under-prediction of seagrass area, especially in many coastal waters of the Bohai Sea, the Yellow Sea, and the South China Sea (Figs.3A–3D). Importantly, the predict- ed distribution areas for nearly all families showed a de- creasing tendency by 2100 (Figs.3K–3N). Nevertheless, the map derived from seagrasses had a larger predicted distri- bution area, with an increasing tendency by 2100 (Fig.3E and 3O).

Table 4 MaxEnt receiver operating characteristic curve (AUC), probability of distribution, and predicted areas of seagrasses

Fig.3 MaxEnt model predicted environmental range for seagrasses in Chinese coastal waters.

3.3 Centroid Migration of Suitable Habitat for Seagrass

Except for Hydrocharitaceae, all the other families tend to migrate northwest in the future. For individual families, Cymodoceaceae migrated nearly 200km, which was the one that might move the longest distance. When modeled using all seagrass species, we obtained a relatively consis- tent result. The centroid of suitable habitat would shift slight- ly to the northwest (Fig.4).

4 Discussion

Researchers and managers have been calling for increased frequency and accuracy in the mapping of seagrass distri- bution to benefit seagrass conservation for over a decade (Green and Short, 2003; Bittner., 2020). As mention- ed above, Chinese researchers have exerted remarkable ef- forts on seagrass ecosystems conservation and restoration in recent years (Xu., 2021). A high-accuracy nation- wide seagrass distribution map will promote the manage- ment and protection of seagrass in China. Basing on ma- rine environmental datasets and seagrass occurrence data, we attempted to predict the potential distribution of known seagrass species in Chinese coastal waters and reveal the past and future trends based on the MaxEnt model. How- ever, we used the global datasets as a proxy because of the inadequacy of Chinese marine environmental data and sea- grass monitoring data.

For the present, the predicted distribution area was far beyond the distribution range of seagrasses known in Chi- na (Zheng., 2013). The basic monitoring is not well developed, and previous studies may have greatly under- estimated the distribution area of seagrasses in China. In addition, many new seagrass meadows have been discover- ed in several provinces in recent years (Zhang., 2019). In terms of the potential distribution area, our model was consistent with the known distribution of seagrasses in Chi- na, such as the specific distribution of Cymodoceaceae in Hainan and Taiwan and the distribution of Zosteraceae along the northern coast of China (Zheng., 2013). However, some hotspots of seagrass were missed in our model, such as the Miaodao Islands in the Shandong Peninsula for Zos- teraceae (Wang., 2019). In the present study, all the models showed high AUC values, which indicated good prediction accuracies. However, similar to the results of Jayathilake and Costello (2018), the individual family models could under-predict the distribution of seagrass owing to the variations in species environmental niches, and the model using all species records provided the most spatial- ly accurate map. For instance, distribution models can be strongly influenced by records. Additional data directly fromChinese coastal areas will greatly improve the accuracy of SDMs.

Fig.4 Changes in suitable ranges of seagrass projected by the MaxEnt model. The arrows and the values show direction and distance between distribution centroids. Direction is measured in degrees north by west, and the blue, green, and red dots indicate distribution centroids of the past, present, and future, respectively.

Physical variables (., temperature, salinity, depth, wave height, and photosynthetically active radiation) and chemi- cal parameters (, pH and nutrients such as phosphate, nitrate, and dissolved oxygen concentration) control the dis-tribution of seagrasses (Waycott., 2009; Jayathilakeand Costello, 2018). Our analysis of variable contributions showed that the physical ones (., depth, land distance, and sea surface temperature) had the greatest contribution in predicting seagrass distributions, regardless of families or seagrasses, which was consistent with the results of otherstudies (Jayathilake and Costello, 2018; Bittner., 2020). However, these results were obviously diverse in different spatial scales for different seagrass species when consider- ing a single variable. For instance, benthic light availabi- lity represented one of the most influential predictors in the Texas Gulf Coast (Bittner., 2020), whereas maxi- mum sea surface temperature had the greatest contribution in predicting global species distributions (Jayathilake and Costello, 2018). Importantly, our model clearly showedthat seagrass is likely to appear in shallow areas (Staehr., 2019), where light conditions are within the optimal range for this species (Krause-Jensen., 2021).

In addition to physical variables, nutrient concentrationsranked among the key influential predictors for habitat sui- tability in our study area because nutrient availability con- trols seagrass growth, distribution, and metabolism (Papa- ki., 2020; Leiva-Dueñas., 2021). Chlorophyll-concentration also had an important influence on the dis- tribution of seagrasses, especially for Cymodoceaceae. Chlo- rophyll-concentration is the key parameter determining seagrass presence/absence in certain areas. However, someissues cannot be ignored. For instance, sediment conditions are critical for determining the possible coverage of sea- grass (Staehr., 2019; Stankovic., 2021). Sediment properties were not included in the dataset used in thisstudy, which may have influenced the model results to some extent.

Accurate and effective evaluation of suitable habitats for species is essential for the identification and determination of priority protected areas (Rangel., 2018). Modeling can be employed to estimate the spatial distribution of po- tential habitat areas, where conditions in the past or in the future can be estimated. In recent years, SDMs have be-come an indispensable tool to describe the complex ecolo- gical relationships between species and their environment and to predict a distribution over multiple spatial and tem- poral scales. Meanwhile, SDMs allow us to identify the po- tential distribution range of seagrasses under changing con- ditions and to explore the impact of global climate change. MaxEnt (Phillips., 2006) uses presence-only data and outperforms other algorithms when applied to small data- sets (de la Hoz., 2019). However, few studies focused on the potential distribution prediction of marine organisms in China. A good survey of the current status of marine biological distribution, integration of various biological andenvironmental datasets, and in-depth research on species distribution models will lay a theoretical and scientific foundation for coastal biological resources and ecological and environmental protection.

On the one hand, a correlation analysis was conducted to remove highly correlated variables, thereby improving modeling accuracy and avoiding common mistakes. In ad- dition, variables with a permutation importance of 0 were deleted in the final models. For SDMs, the variable choiceis important because inadequate variables reduce model ac-curacy and affect subsequent conclusions. For different fa- milies and seagrasses, the environmental variables used in the models were dissimilar and ultimately affected the si- mulation results. On the other hand, the environmental va- riables were re-interpolated to obtain higher spatial reso- lution. The availability of user-friendly, high-resolution en- vironmental datasets is important for distribution model- ing in aquatic environments (Assis., 2018). On the ba- sis of the study by Jayathilake and Costello (2018), the spa- tial resolution of layers used in our study was 30 arcsec, which is reasonable for the representation of seagrass dis- tributions at regional scales, especially when calculating the distribution areas.

The temporal resolution of environmental layers is an- other issue to be considered, especially when paying atten- tion to dynamic patterns of distributions. In this study, the dataset used was averaging environmental variables over large time periods (Basher., 2018), which may mask the underlying dynamic patterns and produce a less realis- tic model (Fernandez., 2017). Several studies have suggested that the use of contemporaneous environmental data, such as daily or weekly fields, is preferable to fitting and projecting models on coarse-scale climatological fields,particularly in highly dynamic domains (Forney., 2012). Biologically relevant time scales can vary from thousands of years to minutes, indicating the need to elucidate how different temporal scales might affect SDMs in the marine realm (Fernandez., 2017).

Climate change is a growing concern for seagrass eco- systems (Short., 2016; Edwards, 2021). Obvious evi- dences and previous results showed that dramatic altera- tions driven by global climate change have occurred in oceans, and such changes are the principal cause of sea- grass ecosystem degradation (Wernberg., 2016; Perry., 2019). Overall, the potential distribution of seagrass varies in different regions in the future. For instance, Che- faoui. (2018) indicated a dramatic loss of seagrass ha- bitat under projected climate change in the Mediterranean Sea. By contrast, our results showed an increasing trend for seagrass in Chinese coastal waters by 2100. Although li- mited by the quality of environmental datasets, the presentstudy still provides a scientific basis for the dynamic changes in seagrass distribution under global climate change. In- terestingly, the suitable habitat for almost all seagrass spe- cies will shift northwest in the future despite the differencein travel distances. This finding might be evidence that sea- grass meadows are moving to the polar or deep sea in re- sponse to global warming.

The distribution limit of seagrass changed tightly in Chi- na under climate change, especially for Zosteraceae. In spe- cific, the southern limit of Zosteraceae might move 300km to the north in the future. Similarly, Valle. (2014) indicated that increasing seawater temperature would trig- ger a poleward shift ofhabitat suitability.Meanwhile, Franssen. (2011) proposed transcriptomic resilience to global warming in the seagrass. This phenomenon did not appear when other fami- lies or all seagrasses were used in modeling. Many stud- ies have indicated that ocean warming will continue to drivelatitudinal abundance shifts in marine species (Hastings., 2020; Trisos., 2020; Nguyen., 2021). Fur- ther work is warranted to elucidate the impacts of global climate change on seagrass distribution.

Awareness of the presence of seagrass meadows is thefirst step for protection, restoration, and sustainable ma-nagement (Staehr., 2019; Unsworth., 2019). Com- bined withinvestigation, this model provides a tool to identify the potential regional occurrence of seagrass at a national scale and highlight the areas where restoration efforts are likely to be successful (McKenzie., 2020). This modeling technique is also affected by autocorrela- tion because species presence sampling is inherently biased(Townhill., 2018). Consequently, identifying where sea- grasses do not occur is also very important because many areas might have not been observed,., vast areas of South East Asia (McKenzie., 2020). Finally, our mo- del will be strengthened when high-quality data layers (both for marine environment datasets and seagrass occurrence data in local areas) are available and will benefit from any advances in ecological theory, statistical methods, and in- creases in computational power.

5 Conclusions

This study is the first to explore the potential distribu- tion of seagrass meadows in China by using species distri- bution models and examine their shift trends under global climate change. Results showed that the potential distribu- tion of seagrass is much larger than currently known dis- tributions, suggesting that seagrass monitoring and protec- tion still have a long way to go. Equally noteworthy is thatthe suitable habitat for seagrasses will shift northwest inthe future, which will provide fresh insights for develop-ing management policies and conservation strategies for seagrasses according to the climate changes. Aside from high modeling accuracy, the lack of high-quality seagrass occurrence data and marine environment data is still a key constraint for accurately predicting seagrass distribution in China.

Acknowledgements

This work was supported by the National Key R&D Pro- gram of China (No. 2019YFC1408405-02), the National Natural Science Foundation of China (No. 6207070555), and the Youth Foundation of the Shandong Academy of Sci- ences (No. 2019QN0026).

Adams, M. P., Koh, E. J., Vilas, M. P., Collier, C. J., Lambert, V. M., Sisson, S. A.,., 2020. Predicting seagrass declinedue to cumulative stressors., 130: 104717.

Assis, J., Tyberghein, L., Bosch, S., Verbruggen, H., Serrão, E. A., and De Clerck, O., 2018. Bio-ORACLE v2.0: Extend- ing marine data layers for bioclimatic modelling., 27 (3): 277-284.

Basher, Z., Bowden, D. A., and Costello, M. J., 2018. GMED: Global Marine Environment Datasets for environment vi- sualisation and species distribution modelling., 2018: 1-62.

Bittner, R. E., Roesler, E. L., and Barnes, M. A., 2020. Using species distribution models to guide seagrass management., 240: 106790.

Chefaoui, R. M., Duarte, C. M., and Serrão, E. A., 2018. Dra- matic loss of seagrass habitat under projected climate change in the Mediterranean Sea., 24 (10): 4919-4928.

Cullen-Unsworth, L. C., and Unsworth, R., 2018. A call for seagrass protection., 361 (6401): 446-448.

de la Hoz, C. F., Ramos, E., Puente, A., and Juanes, J. A., 2019.Temporal transferability of marine distribution models: The role of algorithm selection., 106: 105499.

Duarte, C. M., Losada, I. J., Hendriks, I. E., Mazarrasa, I., and Marbà, N., 2013. The role of coastal plant communities for climate change mitigation and adaptation., 3 (11): 961-968.

Edwards, A. J., 2021. Impact of climatic change on coral reefs, mangroves, and tropical seagrass ecosystems. In::. Eisma, D., ed., CRC Press, Boca Raton, 209-234.

Fernandez, M., Yesson, C., Gannier, A., Miller, P. I., and Aze- vedo, J. M., 2017. The importance of temporal resolution for niche modelling in dynamic marine environments., 44 (12): 2816-2827.

Ferrari, R., Malcolm, H., Neilson, J., Lucieer, V., Jordan, A., Ingleton, T.,., 2018. Integrating distribution models and habitat classification maps into marine protected area plan- ning., 212: 40-50.

Forney, K. A., Ferguson, M. C., Becker, E. A., Fiedler, P. C., Redfern, J. V., Barlow, J.,., 2012. Habitat-based spatial models of cetacean density in the eastern Pacific Ocean., 16 (2): 113-133.

Franssen, S. U., Gu, J., Bergmann, N., Winters, G., Kloster- meier, U. C., Rosenstiel, P.,., 2011. Transcriptomic re- silience to global warming in the seagrass, a marine foundation species., 108 (48): 19276-19281.

Global Biodiversity Information Facility (GBIF), 2017. Ac- cessed at http://www.gbif.org/species (in 2020).

Gumusay, M. U., Bakirman, T., Tuney Kizilkaya, I., and Aykut, N. O., 2019. A review of seagrass detection, mapping and monitoring applications using acoustic systems., 52 (1): 1-29.

Hastings, R. A., Rutterford, L. A., Freer, J. J., Collins, R. A., Simpson, S. D., and Genner, M. J., 2020. Climate change drives poleward increases and equatorward declines in ma- rine species., 30 (8): 1572-1577.

Horton, T., Kroh, A., Ahyong, S., Bailly, N., Boyko, C. B.,Brandão, S. N.,., 2020. World Register of Marine Spe- cies. Available from http://www.marinespecies.org at VLIZ. Accessed 2020-09-25.

Houngnandan, F., Kéfi, S., and Deter, J., 2020. Identifying key- conservation areas forseagrass beds., 247: 108546.

Jayathilake, D. R., and Costello, M. J., 2018. A modelled glo- bal distribution of the seagrass biome., 226: 120-126.

Jiang, Z., Cui, L., Liu, S., Zhao, C., Wu, Y., Chen, Q.,., 2020. Historical changes in seagrass beds in a rapidly ur- banizing area of Guangdong Province: Implications for con- servation and management., 22: e01035.

Kendrick, G. A., Orth, R. J., Statton, J., Hovey, R., Ruiz Mon- toya, L., Lowe, R. J.,., 2017. Demographic and gene- tic connectivity: The role and consequences of reproduction, dispersal and recruitment in seagrasses., 92 (2): 921-938.

Krause-Jensen, D., Duarte, C. M., Sand-Jensen, K., and Carstensen, J., 2021. Century-long records reveal shifting chal- lenges to seagrass recovery., 27 (3): 563-575.

Larkum, A. W., Waycott, M., and Conran, J. G., 2018. Evolu- tion and biogeography of seagrasses. In:. Springer, Cham, 3-29.

Leiva-Dueñas, C., Cortizas, A. M., Piñeiro-Juncal, N., Díaz- Almela, E., Garcia-Orellana, J., and Mateo, M. A., 2021. Long-term dynamics of production in western Mediterra- nean seagrass meadows: Trade-offs and legacies of past dis- turbances., 754: 142117.

Martínez, B., Radford, B., Thomsen, M. S., Connell, S. D., Carreño, F., Bradshaw, C. J.,., 2018. Distribution mo- dels predict large contractions of habitat-forming seaweeds in response to ocean warming., 24 (10): 1350-1366.

McKenzie, L. J., Nordlund, L. M., Jones, B. L., Cullen-Un- sworth, L. C., Roelfsema, C., and Unsworth, R. K., 2020. The global distribution of seagrass meadows., 15 (7): 074041.

Meng, W., Feagin, R. A., Hu, B., He, M., and Li, H., 2019. The spatial distribution of blue carbon in the coastal wet- lands of China., 222: 13-20.

Nguyen, H. M., Ralph, P. J., Marín-Guirao, L., Pernice, M., and Procaccini, G., 2021. Seagrasses in an era of ocean warm- ing: A review., 96 (5): 2009-2030.

Olsen, J. L., Rouzé, P., Verhelst, B., Lin, Y. C., Bayer, T., Col- len, J.,., 2016. The genome of the seagrassreveals angiosperm adaptation to the sea., 530 (7590): 331-335.

Olsen, Y. S., Collier, C., Ow, Y. X., and Kendrick, G. A., 2018. Global warming and ocean acidification: Effects on Austra- lian seagrass ecosystems. In:. Springer, Cham, 705-742.

Perry, D., Staveley, T., Deyanova, D., Baden, S., Dupont, S., Hernroth, B.,., 2019. Global environmental changes ne- gatively impact temperate seagrass ecosystems., 10 (12): e02986.

Phillips, S. J., Anderson, R. P., and Schapire, R. E., 2006. Ma- ximum entropy modeling of species geographic distributions., 190 (3-4): 231-259.

Phillips, S. J., Anderson, R. P., Dudík, M., Schapire, R. E., and Blair, M. E., 2017. Opening the black box: An open-source release of Maxent., 40 (7): 887-893.

Rangel, T. F., Edwards, N. R., Holden, P. B., Diniz-Filho, J. A. F., Gosling, W. D., Coelho, M. T. P.,., 2018. Modeling the ecology and evolution of biodiversity: Biogeographical cradles, museums, and graves., 361 (6399): eaar5452.

Schultz, S. T., 2008. Seagrass monitoring by underwater vi- deography: Disturbance regimes, sampling design, and sta- tistical power., 88 (3): 228-238.

Short, F. T., Coles, R., Fortes, M. D., Victor, S., Salik, M., Is- nain, I.,., 2014. Monitoring in the Western Pacific re- gion shows evidence of seagrass decline in line with global trends., 83 (2): 408-416.

Short, F. T., Kosten, S., Morgan, P. A., Malone, S., and Moore, G. E., 2016. Impacts of climate change on submerged and emergent wetland plants., 135: 3-17.

Short, F. T., Polidoro, B., Livingstone, S. R., Carpenter, K. E., Bandeira, S., Bujang, J. S.,., 2011. Extinction risk as- sessment of the world’s seagrass species., 144 (7): 1961-1971.

Staehr, P. A., Göke, C., Holbach, A. M., Krause-Jensen, D., Timmermann, K., Upadhyay, S.,., 2019. Habitat mo- del of eelgrass in Danish coastal waters: Development, va- lidation and management perspectives., 6: 175.

Stankovic, M., Hayashizaki, K. I., Tuntiprapas, P., Rattana- chot, E., and Prathep, A., 2021. Two decades of seagrass area change: Organic carbon sources and stock., 163: 111913.

Telesca, L., Belluscio, A., Criscoli, A., Ardizzone, G., Aposto- laki, E. T., Fraschetti, S.,., 2015. Seagrass meadows () distribution and trajectories of change., 5 (1): 1-14.

Townhill, B. L., Tinker, J., Jones, M., Pitois, S., Creach, V., Simpson, S. D.,., 2018. Harmful algal blooms and cli- mate change: Exploring future distribution changes., 75 (6): 1882-1893.

Trisos, C. H., Merow, C., and Pigot, A. L., 2020. The project- ed timing of abrupt ecological disruption from climate change., 580 (7804): 496-501.

Unsworth, R. K., McKenzie, L. J., Collier, C. J., Cullen-Un- sworth, L. C., Duarte, C. M., Eklöf, J. S.,., 2019. Glo- bal challenges for seagrass conservation., 48 (8): 801- 815.

Unsworth, R. K., McKenzie, L. J., Nordlund, L. M., and Cul- len-Unsworth, L. C., 2018. A changing climate for seagrass conservation?, 28 (21): R1229-R1232.

Unsworth, R. K., Nordlund, L. M., and Cullen-Unsworth, L. C., 2019. Seagrass meadows support global fisheries pro- duction., 12 (1): e12566.

Valle, M., Chust, G., del Campo, A., Wisz, M. S., Olsen, S. M., Garmendia, J. M.,., 2014. Projecting future distri- bution of the seagrassunder global warming and sea level rise., 170: 74-85.

Wang, M., Zhang, H., and Tang, X., 2019. Growth characte- ristics of a restoredpopulation in the Shan- dong Peninsula, China: A case study., 144: 122-132.

Wang, M. X., Chen, X. Y., Zhang, J., Song, Y. H., and Yang, J., 2021. Biodiversity of Chordata in the Philippine Sea: A case study based on OBIS., 29: 1481-1489.

Waycott, M., Duarte, C. M., Carruthers, T. J., Orth, R. J., Den- nison, W. C., Olyarnik, S.,., 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems., 106 (30): 12377-12381.

Wernberg, T., Bennett, S., Babcock, R. C., De Bettignies, T., Cure, K., Depczynski, M.,., 2016. Climate-driven re- gime shift of a temperate marine ecosystem., 353 (6295): 169-172.

Xu, S., Qiao, Y., Xu, S., Yue, S., Zhang, Y., Liu, M.,., 2021. Diversity, distribution and conservation of seagrass in coastal waters of the Liaodong Peninsula, North Yellow Sea, northern China: Implications for seagrass conserva- tion., 167: 112261.

Yu, S., Liu, S., Jiang, K., Zhang, J., Jiang, Z., Wu, Y.,., 2018. Population genetic structure of the threatened tropical seagrassin Hainan Island, China., 150: 64-70.

Zhang, X., Lin, H., Song, X., Xu, S., Yue, S., Gu, R.,., 2019. A unique meadow of the marine angiosperm, covering a large area in the turbid intertidal Yel- low River Delta, China., 686: 118-130.

Zheng, F., Qiu, G., Fan, H., and Zhang, W., 2013. Diversity, distribution and conservation of Chinese seagrass species., 21 (5): 517.

April 6, 2021;

June 3, 2021;

October 26, 2021

© Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2022

. Tel: 0086-532-83591310

E-mail: mwang@qnlm.ac

(Edited by Qiu Yantao)