INTRODUCTION
The physical structure of a coral reef has profound effects on nearshore hydrodynamics which in turn affects ecosystem functioning and, ultimately, its associated biodiversity. Structurally complex habitats facilitate the coexistence of species through niche partitioning and the provision of shelters from predators and environmental stressors (Warfe and Barmuta, 2006; Willis, Winemiller, and Lopez-Fernandez, 2005). Therefore, the loss of architectural complexity can cause reduction in biodiversity, and thus, compromise habitat productivity (Beukers and Jones, 1997). Qualitatively, roughness or rugosity is a measure of a surface's departure from one that is geometrically smooth, which is a parameter that has been used to quantify topographic complexity of coral reefs (Hearn, 2011). Turbulence induced by the coral colonies affects the rate of particle entrainment in the water column. In consequence, food acquisition for several reef species depends on the dispersion processes and how near-bed flow dynamics exchange mass between the reef and the surrounding ocean. As the coral reef grows and becomes rougher, the assemblages of benthic organisms that can be supported by it increase in terms of density and species richness (Falter, Atkinson, and Coimbra, 2005; Monismith, 2007; Reidenbach et al., 2006). Loss of roughness also has important implications for coastal protection, since large amounts of wave energy are dissipated by reef crests due their high rugosity (Lowe et al., 2005; Rogers et al., 2016).
In the Caribbean, coral degradation has occurred for the last 40 years (Alvarez-Filip et al., 2009), through a combination of human (e.g., overfishing, sedimentation, eutrophication) and natural disturbances (Gardner et al., 2003; Hughes, 1994; Hughes et al., 2005; Suchley and Alvarez-Filip, 2018). Natural disturbances may be physical (e.g., hurricanes) or biological (e.g., diseases). The effects of hurricanes on coral reefs have been widely studied and include the mechanical stress produced by storm surges and blows suffered by coral fragments (Woodley et al., 1981); and the abrasive effect of sediment that is easily resuspended and transported during a storm (Hubbard and Pockock, 1972). It has been observed that the recovery of coral reefs from the impact of a hurricane can take decades, or sometimes it would not happen at all. For example, in Puerto Morelos, Mexico, the coral reef community showed no recovery of hard corals 15 years after Hurricane Gilbert (Rodríguez-Martínez et al., 2010). Therefore, hurricanes have a prominent role in controlling the configuration and composition of fringing reefs in the tropical Western-Atlantic (Blanchon et al., 2017; Perry, 2001).
Disturbances tend to synchronize, and single causes for reef degradation are rarely the case. For example, in the Caribbean apart from hurricane effects, this ecosystem has been strongly impacted by the white-band disease, and by a reduction of herbivores such as the black sea urchin (Diadema antillarium) due to overfishing (Alvarez-Filip et al., 2009; Gardner et al., 2003; Hughes, 1994). In the past, many studies have indicated overfishing of herbivorous fish as the key factor for the widespread dominance of macroalgae in coral reef environments, but recent studies suggest that watershed pollutants can now be identified as the main inducers of coral phase shifts, which represents a change in paradigm for coral reef science (Arias-González et al., 2017; Suchley, McField, and Alvarez-Filip, 2016).
In the Mexican Caribbean, continental waters enter the ocean mainly through Submarine Groundwater Discharges (SGD), which represent an important pathway for nutrients and other contaminants that can potentially affect coral reefs (Burnett et al., 2006; Paytan et al., 2006). In karstic regions, the risk of groundwater pollution is high (Lapointe, O'Connell, and Garrett, 1990) because of the rapid recharge and channelized flow pathways through fractures and cave systems (Null et al., 2014). Nutrients promote the growth of macroalgae, which compete with corals for space, reducing their recruitment, growth, fecundity, and survival through various mechanisms (Chadwick and Morrow, 2011; Suchley and Alvarez-Filip, 2018). Therefore, there is a need to better understand how anthropogenic activity and groundwater delivery are linked to land use and pollutant transport in areas with karst geology, since the impact to local coastal ecosystems can have serious detrimental consequences (Null et al., 2014).
The aim of this investigation is to study the link between macroalgae cover and roughness in a reef lagoon that has a spatially varying influence of SGD close to the reefs. The negative relationship between macroalgae cover and reef roughness in regions of high SGD influence would be supporting the idea that phase shifts (from coral to macroalgae dominance in benthic cover) are being intensified by poor water quality in the region.