Wetlands https://doi.org/10.1007/s13157-018-1028-3 APPLIED WETLAND SCIENCE Fine-Scale Mapping of Coastal Plant Communities in the Northeastern USA Maureen D. Correll1,2 & Wouter Hantson1 & Thomas P. Hodgman3 & Brittany B. Cline1,4 & Chris S. Elphick 5 & W. Gregory Shriver4 & Elizabeth L. Tymkiw4 & Brian J. Olsen1 Received: 12 November 2017 /Accepted: 16 March 2018 # Society of Wetland Scientists 2018 Abstract Salt marshes of the northeastern United States are dynamic landscapes where the tidal flooding regime creates patterns of plant zonation based on differences in elevation, salinity, and local hydrology. These patterns of zonation can change quickly due to both natural and anthropogenic stressors, making tidal marshes vulnerable to degradation and loss. We compared several remote sensing techniques to develop a tool that accurately maps high- and low-marsh zonation to use in management and conservation planning for this ecosystem in the northeast USA. We found that random forests (RF) outperformed other classifier tools when applied to the most recent National Agricultural Imagery Program (NAIP) imagery, NAIP derivatives, and elevation data between coastal Maine and Virginia, USA. We then used RF methods to classify plant zonation within a 500-m buffer around coastal marsh delineated in the National Wetland Inventory. We found mean classification accuracies of 94% for high marsh, 76% for low marsh zones, and 90% overall map accuracy. The detailed output is a 3-m resolution continuous map of tidal marsh vegetation communities and cover classes that can be used in habitat modeling of marsh-obligate species or to monitor changes in marsh plant communities over time. Keywords High marsh . NAIP . Random Forest . Remote sensing . Spartina . Tidal marsh Introduction globe. These marshes serve as a gateway between land and sea for humans and wildlife alike, act as a buffer against coastal Coastal marshes are among the world’smostproductiveeco- storms, and provide critical nutrients to marine food webs systems and provide significant services to humans across the (Barbier et al. 2011). Tidal marshes also support and protect biodiversity by providing habitat to marine and estuarine fish, crustacean populations, and migratory birds (Boesch and Electronic supplementary material The online version of this article Turner 1984;Master1992; Brown et al. 2002). (https://doi.org/10.1007/s13157-018-1028-3) contains supplementary ’ material, which is available to authorized users. Within the world s tidal marsh systems, those located along the Atlantic coast of the United States support the * Maureen D. Correll highest number of terrestrial vertebrate specialists de- [email protected] scribed worldwide (Greenberg et al. 2006). This suite of species includes herpetofauna and mammals, but the ma- 1 School of Biology and Ecology, The University of Maine, jority of described vertebrate specialists are birds. Several Orono, ME 04469, USA species are limited completely to these marshes during the 2 Bird Conservancy of the Rockies, Fort Collins, CO 80521, USA breeding season, several of which are in decline (Correll et al. 2017), with extinction predicted for the saltmarsh 3 Maine Department of Inland Fisheries and Wildlife, Bangor, ME 04401, USA sparrow within 50 years (Correll et al. 2017;Fieldetal. 2017a, c). 4 Department of Entomology and Wildlife Ecology, The University of Delaware, Newark, DE 19716, USA These declining species nest predominantly within the high- marsh zone, one of several vegetation communities found with- 5 Department of Ecology and Evolutionary Biology and Center for Conservation and Biodiversity, University of Connecticut, in coastal marshes. High marsh differs from other marsh areas Storrs, CT 06269, USA in elevation, salinity, and frequency of inundation (Bertness Wetlands and Ellison 1987; Pennings and Callaway 1992; Ewanchuk The physical and biological characteristics that differentiate and Bertness 2004) and is characterized by flooding during high marsh from low marsh and other marsh plant communi- spring tides linked to the lunar cycle. In the northeastern ties are potentially detectible using remotely-sensed multi- United States, the plant species Spartina patens, short-form spectral and hyperspectral imagery. Both types of imagery S. alterniflora, Distichlis spicata,andJuncus gerardii charac- can record wavelengths of light outside of the visible range terize high-marsh zones, which also include Salicornia spp., for humans, with hyperspectral imagery recording reflectance Glaux maritima,andSolidago sempervirens (Nixon and values in much finer detail and precision (hundreds of indi- Oviatt 1973;Bertness1991; Emery et al. 2001, Ewanchuk vidual bands recorded) than multispectral imagery (several and Bertness 2004). Conversely, low marsh is characterized wide-ranging bands recorded, e.g. red green blue, or RGB, by daily tidal flooding and is a near monoculture of tall form imagery). Several studies have previously demonstrated dis- S. alterniflora. The surrounding terrestrial border experiences tinct spectral differences between tidal marsh species using infrequent inundation by salt water during extreme tides and hyperspectral imagery (Rosso et al. 2005; Belluco et al. storms, and is characterized by a more diverse flora that is often 2006; Yang 2009). Such imagery, when combined with eleva- dominated by Iva frutescens and Typha spp. (Miller and Egler tion data, has previously produced high-accuracy classifica- 1950; Ewanchuk and Bertness 2004). Introduced Phragmites tions of tidal marsh vegetation communities, albeit at smaller australis (hereafter Phragmites) also occurs within this ecosys- spatial scales (e.g. Hladik et al. 2013). tem, especially around the borders of disturbed marshes Hyperspectral imagery is costly, however, especially across (Chambers et al. 1999; Philipp and Field 2005). large landscapes (Adam et al. 2010); Belluco et al. (2006) These plant community zones can be quickly altered by compared several aerial and satellite sensors with changing both natural and anthropogenic stressors such as sea-level rise, spatial and spectral resolution, and although hyperspectral nutrient run-off from adjacent uplands, and the spread of in- imagery performed slightly better than the multispectral, troduced species (Day et al. 2008). The significant increase in spatial resolution was the most important factor in classifier sea level during recent decades poses one of the largest threats performance. Belluco et al. (2006)recommendtheuseof to these marsh ecosystems. As sea levels encroach on the multispectral satellite imagery for the mapping of marsh veg- marshes’ seaward side and upland marsh migration is limited etation. Beside the visible spectrum (RGB), multispectral im- by human-developed coastal infrastructure (Field et al. 2017b) agery should also include infrared (IR) reflectance values to and upland habitats (Field et al. 2016), a Bpinching effect^ can allow differentiation of vegetation types, calculation of vege- occur, resulting in marsh loss. Coastal marshes can combat tation indices (e.g. the Normalized Difference Vegetation rising sea levels through vertical growth, or accretion Index or NDVI, Rouse et al. 1974) and detection of soil mois- (Kirwan et al. 2016), but when the rate of sea-level rise ex- ture differences (Jin and Sader 2005;Pettorellietal.2005), ceeds the rate of accretion, marsh area will decline (Crosby particularly in tidal wetlands (Klemas 2011). The IR spectrum et al. 2016). Rising sea levels can also drive invasion of high has previously been used as a tool to predict tidal marsh com- marsh areas with flood-tolerant low marsh species (Donnelly munities both in smaller regions within the northeastern and Bertness 2001; Field et al. 2016), causing transition from United States (Gilmore et al. 2008; Hoover et al. 2010; high to low marsh (Kirwan et al. 2016). This pattern, however, Meiman et al. 2012) and elsewhere (Isacch et al. 2006;Liu is not ubiquitous to all marshes (Kirwan and Guntenspergen et al. 2010). An exception to this has been in the classification 2010; Wilson et al. 2014). In addition to sea-level rise, ex- of invasive Phragmites that often borders tidal marshes. treme storm events that flood the coastline have been shown Large-scale classification of this wetland class has met with to permanently alter marsh structure within days (Day et al. some success (e.g. Bourgeau-Chavez et al. 2015,Longetal. 2008) and can have a lasting effect on plant community struc- 2017), however success is limited when using RGB and IR ture and saltmarsh degradation. inputs alone (Samiappan et al. 2017). A large-scale effort to Marsh degradation and rapid change is likely to continue map coastal Phragmites in the northeastern US has not yet into the future due to the paired effects of climate change and been attempted. human development. Sea levels are expected to rise substan- Due to the large spatial scale at which northeastern tidal tially between 2013 and 2100 (IPCC 2014), and continuing marshes occur, publicly-available and low-cost imagery storm events affecting coastal regions are also predicted. The datasets offer the most promising option for repeatedly delin- future distribution of high- and low-marsh habitat therefore eating large swaths of marsh along the coast. Landsat satellite remains uncertain (Chu-Agor et al. 2011;Kirwanetal. imagery provided by the National Aeronautic and Space 2016). It is essential to develop tools to identify coastal marsh Administration