Do mangroves provide an effective barrier to storm surges?

Abstract

(1) conclude that mangroves reduced the death toll from storm surge in the Orissa Super Cyclone. However, it is unclear from their analyses whether or not this effect occurs independently over and above that of other variables known to affect inundation by long-period waves, such as distance from the coast and topography (2). Further, Das and Vincent (1) are wrong to equate storm surges with windgenerated storm waves. Storm surges have a period of hours to days (3) and consequently behave more like the tide or a tsunami. An effect of vegetation has never been questioned (2). The drag of vegetation must reduce wave velocity to some degree. More important questions, however, are how much protection vegetation can provide and how this compares with other mitigating factors (2). The results suggest that mangroves offered little protection (1). The correlation coefficient (r ϭ Ϫ0.13) between mangrove width and village deaths suggests that mangroves explain less than 2% of the variation in deaths. Clearly, other factors were also important. The study site covers only the northern extreme of the area affected by the cyclone and only 254 of Ͼ10,000 deaths occurred here. When the total area is considered, there is clear aggregation of deaths, which were much higher closer to where the storm crossed the coast (Fig. 1). Presumably, village proximity to the storm center or other variables such as distance from the coast and height above sea level were more important than mangroves in determining deaths per village. Indeed, these last two variables are significant whenever included in the models. Why then the emphasis on mangroves when no formal methods of model selection were used (4)?Even accepting the results, do they indicate that mangroves can provide an effective shield against storm surges, as suggested by the Food and Agriculture Organization? Surgerelated deaths were recorded in villages Ͼ2 km inland (1), suggesting the vegetation barrier will need to be very wide in similar low-lying areas. Indeed, in areas where high ground occurs close to the coast, the effect of vegetation is unlikely to be detectable (2). Finally, is revegetation an economical option for disaster mitigation? The authors estimate an opportunity cost of $300,000 (US) for every hectare of mangrove not cleared for farming. In contrast, the cost of an early warning system is minimal, because the existing weather forecasting capacity was sufficient to issue a warning that saved Ϸ5 lives per village (1). We conclude that although mangroves may have reduced deaths in this event (1), the effect is likely to be small, particularly when compared to other variables. Furthermore, the cost of revegetation is high and the effectiveness of a mangrove barrier is low when compared to an early warning system. The value in conserving coastal forests is preventing occupation of the most dangerous areas, which are close to the coast in low-lying areas (5).

Reply to Baird et al.: Mangroves and storm protection: Getting the numbers right

“…Baird et al (1) appear to accept the principal conclusion of our paper (2), that mangroves reduced the number of deaths from the storm surge of the 1999 super cyclone in Orissa. They are unsure whether we controlled for distance to coast and topography, and they imply that we did not control for proximity to storm path.…”

“…For villages within 10 km of the coast, our regression results imply that remaining mangroves reduced the average number of deaths by 69% (Table S9 in ref. 2). This is hardly ''small,'' as Baird et al claim (1).…”

“…Baird et al (1) compound this error by equating the impact of mangroves to the proportion of variation in deaths that mangroves explain. The impact is instead related to the regression parameter on 1999 mangrove width.…”

Influence of coastal vegetation on the 2004 tsunami wave impact in west Aceh

“…The present damage evaluation focused on casualties and structural damage, similar to other studies (2, 4, 5, 11), but it considered criticism that some of these had received (7)(8)(9)(10)18). Issues regarding spatial autocorrelation in assessment methods were addressed through the use of a spatially explicit statistical model.…”

“…Parameters such as stem diameter and height as well as a "bioshield" width were identified as key vegetation characteristics with a bearing on impact mitigation (6). However, several studies advocating bioshields have been criticized for lacking empirical evidence to support the protective function of vegetation, some even suggesting that bioshields may give a false sense of security to coastal populations (7)(8)(9)(10). The role of vegetation in tsunami impact mitigation still remains a controversial issue (11)(12)(13).…”