In recent years, large expanses of Tampa Bay have experienced natural seagrass recolonization correlated with the achievement of improved water quality. However, it now appears that seagrass has slowed its rate of recolonization and recovery does not approach that of the historical, pre-impact distributions of the early part of the 20th Century. One theory is that the substantial reduction of long shore sandbars during the mid to late 20th Century has created a situation where waves are now reaching portions of the shoreline that historically had enjoyed a more quiescent setting suitable to the establishment and maintenance of extensive seagrass habitats. Therefore, we employed a wave exposure model (Relative Exposure Index - REI) designed to evaluate the influence of these bars on seagrass cover. Model results revealed that the bars in the Tampa Bay system, placed as they are on large shoals along the margins of the bay, might not provide substantial wave reduction effects until they are emergent or nearly emergent from the water. Model output also demonstrated the strong effects of the bars on REI reduction, but location of the bar on the shoal (seaward or landward) logically determined the extent of the effect of the bar on the immediate shoal area, and thus, potential seagrass habitat. Logistic regression revealed that REI and water depth were both strongly influential in the prediction of existing seagrass cover and that the greater the water depth and the greater the REI, the lower the probability of seagrass cover. However, in areas where the REI was approximately 4000 or less, the role of wave exposure was predicted to be minimal and may not explain the absence of seagrass and its lack of recovery. On average, sample points located landward of bars were forecast to have a much higher probability of seagrass cover than those seaward of the bar, leading us to conclude that the existence of an emergent bar will produce a forecast of substantially higher
probabilities of seagrass cover in the lee of the bar. The effects of bridges on seagrass abundance, coupled with field observations suggests that bridges should be modeled as partially permeable structures and that their presence in Old Tampa Bay indicates they have had only local influence on increasing seagrass abundance. Therefore, consistent with past hypotheses, loss of bars, particularly on the relatively exposed eastern margin of the Bay are hindcast to have resulted in the loss of large areas of seagrass, although these losses cannot be unequivocally assigned solely to changes in wave climate.