Seagrasses compose the most widespread coastal ecosystems in the world (Green and Short, 2003), providing important ecological services to the marine environment (Costanza and others, 1997) through their high productivity, stability and enrichment of sediments, and support of a complex trophic food web (Duarte and Cebrián, 1996; Hemminga and Duarte, 2000). Seagrass meadows may also reduce the load of disease-causing bacteria, including those that affect humans as well as marine fishes, invertebrates, and mammals, in seawater surrounding seagrass beds (Inaba and others, 2017; Lamb and others, 2017). As such, their distribution and aerial extent may be important indicators of water quality and overall health of a coastal ecosystem (Dennison and others, 1993; Short and Neckles, 1999). Therefore, it is important to inventory seagrass distribution to establish a baseline for future monitoring of trends in these critically important habitats.

Eelgrass (Zostera marina) and widgeongrass (Ruppia maritima) are important seagrasses of protected coastal embayments of the central Pacific coast of Baja California (26 to 28 degrees North; Ibarra-Obando and Rios, 1993; Wyllie-Echeverria and Ackerman, 2003, Ward and others, 2004, Ward 2022a). This region is within the southern edge of the range of eelgrass in the north Pacific (Wyllie-Echeverria and Ackerman, 2003). Here, seagrass meadows provide food and refuge for a variety of ecologically and commercially important fauna, such as bivalves (for examples, oysters [Crassostrea gigas] and scallops [Argopecten ventricosus]), lobsters (Panulirus interruptus), crabs, turtles (Chelonia mydas), juvenile fish, and hundreds of thousands of migrating and resident waterbirds, including greater than (>) 100 species and about 30 percent of the world’s population of black brant (Branta bernicla nigricans; Massey and Palacios, 1994; Page and others, 1997; Aguirre-Muñoz and others, 2001; Ibarra-Obando and others, 2001; Lewis and others, 2020). A significant portion of the economy (for example, commercial and sport fisheries, aquaculture, sport hunting, and tourism) of Baja California depends on the marine resources of these coastal embayments (Aguirre-Muñoz and others, 2001).

Despite the importance of seagrasses to the health of the coastal ecosystems of Baja California, quantifiable data on the spatial distribution of seagrasses are largely lacking. These data are particularly important because coastal areas of Baja California are under increasing threats, both direct and indirect, from human activity (for example, aquaculture, overfishing, and climate warming; Ibarra-Obando and Escofet 1987; Ortega and Castellanos, 1995; Aguirre-Muñoz and others, 2001). Moreover, these threats have been linked to declines in the abundance and spatial extent of seagrasses in northern Baja California (Cabello-Pasini 1984; Ward and others 2003) and in southern California (Johnson and others, 2003, Harenčár and others, 2018). Changes to the distribution and abundance of seagrasses may have major implications for migratory birds and mammals that forage in these habitats (Ward and others, 2005).

The objective of this study was to estimate the distribution and spatial extent of seagrasses in embayments of the central Pacific coast of Baja California, Mexico to acquire baseline data that can be used to monitor trends in this critically important habitat.

Seagrasses composed a significant portion of the spatial extent in five of the six embayments (table 1). In the three largest (>12,000 ha; LOL, LGN and LSI) embayments, seagrasses covered between 34 percent and 47 percent of the total area (table 1). Salt marshes or mangroves (35–67 percent) dominated cover types in the other, much smaller (<4,000 ha; LM, EC, ESJ), embayments. Eelgrass was the most spatially abundant seagrass in all embayments, dominating the low intertidal and shallow subtidal (0.0 to –2.5 m) areas (Ward 2022a, b). Monotypic stands of widgeongrass were predominately in the mid-intertidal (approximately 0.3–1.0 m) whereas mixed stands of eelgrass and widgeongrass were in slightly lower intertidal (approximately 0.0–0.3 m) areas.

In northern embayments, particularly LOL, continuous stands (density >50 percent) of intertidal eelgrass were found in the front half of the lagoon, mainly on the backside of the sandflats at the mouth of the lagoon or mid-lagoon (fig. 2). In the back half of LOL, eelgrass occurred primarily in the subtidal, probably because of the higher water temperatures, reduced turnover of seawater during tidal exchanges, and higher salinities than in the front half of the lagoon (Phleger and Ewing, 1962). Widgeongrass, which tolerates high temperature and salinities (Mayer and Low, 1970), was more abundant in this part of the lagoon than in the front half of the lagoon.

Salt marshes (13–67 percent) and channels (11–33 percent) were the next most abundant cover type in northern embayments (table 1). In LM, the cordgrass-dominated salt marshes composed over half (67 percent) of the overall spatial extent of this lagoon (table 1). The cordgrass-dominated salt marshes of these northern embayments are the largest in Mexico and among the largest stands of this native cordgrass (Spartina foliosa) across its range which extends from northern California to the central Pacific coast of Baja California (Spicher and Josselyn, 1985; Ward and others, 2004). Channels were most significant in LOL with water depths reaching >20 m near the mouth of the lagoon. Because of the overall length (51 km) and width (6–13 km) of LOL, timing of tides can vary by up to 3 hours between the mouth and back of the lagoon (this study).

The overall classification accuracy of the 1999 satellite imagery was good at 79 percent (table 2). The accuracy of all seagrass species combined was also good at 95 percent (table 2); however, when seagrass was split into separate specie categories accuracy declined considerably (accuracy of intertidal eelgrass: 41 percent; intertidal widgeongrass: 23 percent; intertidal seagrass mixed: 67 percent). This drop in accuracy was expected because of the overlap in spectral characteristics and spatial extent of eelgrass and widgeongrass and the difficulty to distinguish spectral signals between these species (Mumby and others, 1997; Ward and others, 2004). This problem of subcover type accuracy was also intensified by the low densities of the seagrasses in these embayments (eelgrass: mean = 20 percent; widgeongrass: mean = 10 percent; Ward, 2022a).

In the southern embayments, widgeongrass dominated high intertidal areas and eelgrass dominated low intertidal and subtidal areas. Eelgrass was more abundant at deeper depths in the southern embayments than in the northern embayments (table 1), possibly due to the increased irradiance and warmer temperatures that plants experience in southern than in northern embayments (Cabello-Pasini and others, 2003). Future research into the influence of warmer temperatures is needed to address this possibility. Overall, however, eelgrass populations of the central Pacific coast of Baja California are exposed to twice the irradiance levels and 18–32 percent higher average maximum water temperatures as eelgrass populations further north along the Pacific coast of Baja California (for example, Bahia San Quintin; Cabello-Pasini and others, 2003).

In LSI, seagrass composed about 34 percent of the total area of the lagoon (table 1; fig. 3). Eelgrass was the dominant seagrass in the lagoon with continuous stands of eelgrass occurring primarily in subtidal areas of the mid to back half of LSI (fig. 3). Overall, eelgrass composed about 63 percent (5,301 ha) of the spatial extent of seagrasses in the lagoon, of which 59 percent were in subtidal areas (table 1). The maximum depth limit of eelgrass in LSI was about –6 m MLLW. In contrast, monotypic stands of widgeongrass composed about 33 percent of the spatial extent of seagrasses and nearly all its coverage occurred in intertidal areas (table 1; fig. 3). Mixed beds of intertidal eelgrass and widgeongrass composed a small portion (4 percent) of the seagrass coverage in LSI. Channels (31 percent) and unvegetated sand and mudflats (23 percent) were the next most abundant cover type in LSI. Salt marshes, which were dominated by pickleweed, and mangroves were of similar size, composing about 6 percent of the overall extent of the lagoon.

Whereas the overall classification accuracy of the photomosaic imagery of LSI was good at 70 percent (table 3), the classification accuracy of the subcover types ranged from poor (intertidal eelgrass; 35 percent; intertidal mixed seagrasses: 36 percent) to good (intertidal widgeongrass: 67 percent; submerged mixed seagrasses: 79 percent). Reasons for the low classification accuracy of the seagrass subcover types in the photomosaic were like those for the satellite imagery of the northern embayments (in other words, spectral wavelength overlap of the two seagrass species combined with their low-density coverage of the seagrass species).

Spatial coverage estimates for ESJ should be considered preliminary because the overall classification accuracy of this imagery was poor at 40 percent (table 4). The main reason for the low classification accuracy for this embayment was due to the lack of identifiable landmarks to obtain quality control points to register this image. There were a high number of locational mismatches between ground-truthing and classified map points, suggesting a registration issue. This issue was emphasized by the poor accuracy estimates for mangroves, which were highly identifiable in the LSI imagery (>90 percent accuracy).

The 4-band, digital multispectral videography of EC was robust at differentiating primary cover types in this embayment (fig. 4); however, the overall classification accuracy of this imagery was poor at 35 percent (table 5). The poor classification accuracy is also likely related to the same registration problems experienced for the ESJ imagery. EC and ESJ are remotely located with no permanent structures or other landmarks near the embayments to obtain quality control points. As with ESJ, estimates of spatial extent of the cover types of EC should be considered preliminary. A reassessment of seagrasses in EC and ESJ could be made after the EC and ESJ images are re-registered with higher quality control points.

This study provides a baseline assessment for the distribution of seagrasses in embayments of the central Pacific coast of Baja California. We provide the first estimates of the spatial extent of eelgrass and widgeongrass in LOL, LGN, LM, and LSI and preliminary estimates for ESJ and EC that can be used to monitor future changes in the distribution of these seagrass species in these embayments. These six embayments contain a significant portion of seagrasses along the Pacific coast of Baja California (Wyllie-Echeverria and Ackerman, 2003; Ward and others, 2004). Eelgrass monitoring is encouraged in Baja California because of declining trends in eelgrass abundance and distribution in some of its west coast embayments (Ward and others, 2003; Rodríguez-Salinas and others, 2010). Northern Mexico is the southern limit of the range of eelgrass in northeastern Pacific (Wyllie-Echeverria and Ackerman, 2003). Eelgrass populations of this region are already under maximal thermal stress and may be at higher risk from impacts of climate change (Short and Neckles, 1999), such as increased risks of sedimentation from coastal flood events, El Niño storm events, disease, and rising sea temperatures and sea level (Harvell and others, 2002; Ward and others, 2003; Muñiz-Salazar and others, 2006; Cai and others, 2014).