Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery

World Development Vol. xx, pp. xxx–xxx, 2017 0305-750X/Ó 2017 Elsevier Ltd. All rights reserved. www.elsevier.com/locate/worlddev

http://dx.doi.org/10.1016/j.worlddev.2017.02.010

Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery SAUDAMINI DAS* Institute of Economic Growth, India Summary. — Restoration of degraded and depleted mangrove habitats and planting of mangroves over coastal mudflats is happening at many places, but there are few studies that evaluate the flow of ecosystem services from these regenerated ecosystems. The state of Gujarat in Western India has planted thousands of hectares of mangroves over the coastal mudflats and, today, the state’s mangrove cover is nearly double that in the 1930s. However, these mangroves have limiting features: for example, these are mostly single-species, Avicenna marina, and are sparse, and lack freshwater supply. Mangroves provide multiple ecosystem services including nursery and habitat services for fish fry that enhances fish growth. This study evaluates the regenerated forests’ contribution to the fishery sector of Gujarat, both inshore, and offshore, using the difference-in-differences technique, and panel regression estimates. Commercial catch data from secondary sources and primary survey diary on the daily catch of artisanal fishermen are used in the analysis. The results show that the planted mangroves have significantly increased the catch of mangrove-dependent fish in both sectors, and that young planted strands contribute nearly one-fourth of the contribution of natural strands. Despite the limiting features, the contribution of the planted mangroves’ nursery ground and habitat service to the fishery sector of Gujarat state is valued at INR36.04 billion (USD0.57 billion) annually. Ó 2017 Elsevier Ltd. All rights reserved. Key words — ecosystem service, fishery, Gujarat, nursery and habitat, regenerated mangrove, value of planted mangroves

1. INTRODUCTION

with the flow of services from natural mangroves. Further, mud flats have been widely used for mangrove planting (Erftemeijer & Lewis, 1999), though in reality, sub-tidal mudflats are inappropriate for mangrove forest restoration, as was evident from a mangrove restoration project in Philippines where the survival rate was very low and the surviving mangroves showed abysmally stunted growth (Lewis, 2010). Mud flats are proven productive ecosystems with high economic and ecological values (Erftemeijer & Lewis, 1999; Naber, Lange, Hatziolos, & UNEP/WCMC, 2008; UNEP, 2005) and reclaiming these habitats for planting mangroves may prove to be a poor resource allocation decision if the flows of ecosystem services from these planted mangroves are found to be inadequate. Such dilemmas also make evaluation of ecosystem services from planted mangroves an important area of research. Though, there are limited evaluations of planted mangroves from the viewpoint of societal benefits, the

Mangrove forests provide many ecosystem services that increase the welfare of both local and global consumers, like protection to lives and property during coastal disasters, enhancement of fisheries, promotion of biodiversity as mangroves are habitats to numerous flora and fauna, climate control through carbon sequestration, waste processing, food production, recreation, etc. (Aburto-Oropeza et al., 2008; Barbier et al., 2008, 2011; Blaber, 2007; Das & Cre´pin, 2013; Das & Vincent, 2009; MEA, 2005; Meyfroidt & Lambin, 2009; Mukherjee et al., 2014; Valiela, Bowen, & York, 2001). After 1950s, the world witnessed rapid mangrove loss due to various reasons like overharvesting, clearing for developmental uses, or for other high-yielding land uses like aquaculture, agriculture, tourism, etc. (FAO, 2008). However, the rates of mangrove loss have slowed down—the latest estimate is that the world lost 0.19 million hectares of mangroves during 2001–12, much less than the 3.09 million hectares lost during 1980–2000 (FAO, 2008). Based on these data, the annual rates of mangrove loss are also seen to be declining steadily—1.04% during 1980–90 to 0.72% during 1990–2000, and then from 0.66% during 2000–05 to 0.13% during 2001–12. 1 In recent years, probably with an increase in environmental knowledge and awareness on mangrove values, there has been a revival of mangrove forests in many parts of the world—either through ecological restoration of degraded mangrove areas or mangrove planting over non-mangrove areas like mud flats, salt marshes, or degraded coastal lands (like rejected aquaculture ponds) (Field, 1999; Lewis, 2001, 2009). Such policies are also partly instigated by global policy commitments such as the Convention on Biological Diversity. 2 However, there are limited studies that evaluate the flow of ecosystem services from planted mangroves and compare them

* This study was undertaken under The Economics of Ecosystems and Biodiversity (TEEB)—India Initiative program of the Ministry of Environment, Forest and Climate Change (MOEFCC), Government of India, and financed by GIZ (giz│Deutsche Gesellschaft fu¨r Internationale Zusammenarbeit (GIZ) GmbH) Germany. I sincerely thank Shri HK Pandey, Additional Secretary, MOEFCC, GIZ, and the Scientific and Technical Advisory Group members of TEEB India Initiative for this opportunity. I acknowledge the academic and other help received from M Troell, A Thivakaran, E. Vivekandan, J R Bhatt, G Kadekodi, I M Ishwar, T Johri, and A Das and comments received during the 17th annual BIOECON conference where this paper was presented. I also extend my gratitude to the reviewers for their in-depth comments and suggestions. The study findings are the author’s responsibility and do not reflect the opinion of the Government of India. Final revision accepted: February 8, 2017. 1

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

2

WORLD DEVELOPMENT

success of mangrove plantation has been evaluated through ecological parameters like 1. biodiversity richness—composition of microbes, fungi, plants, tropical guilds as well as invertebrates and mud crab populations (Ellison, 2008; Walton, LeVay, Lebata, Binas, & Primavera, 2007); 2. the composition of forest structure through vegetation cover and height, woody density, biomass, basal area, or litter structure (Bosire et al., 2008; Kairo, Lang’at, Dahdouh-Guebas, Bosire, & Karachi, 2008; Macintosh, Ashton, & Havanon, 2002; McKee & Faulkner, 2000); and 3. biotic and abiotic features like soil pH level, organic content, or moisture content between planted and natural mangroves (Khayat & Jones, 1999; Walters, 2000), etc. This study attempts a socio-economic evaluation by measuring the contribution of planted mangroves to the fishery sector of the state of Gujarat in India. The state has successfully planted thousands of hectares of mangroves over coastal mudflats and a further 810 sq km have been identified where mangroves can be planted in future (Pandey, Pandey, & Khokhariya, 2012). Mangrove plantation started in 1948 in the state (Gazetteers, 1971) as a pure public sector activity; it is now being pursued under the public–private–partnership (PPP) model. The flow of ecosystem services from regenerated mangroves is argued to depend on multiple factors like slope and height of mud substratum, distribution of freshwater inputs, species composition, abundance, and size structure of mangrove stands, density of detritivorous invertebrates, energy flows, vertical zonation pattern of organisms, etc. (Kaly & Jones, 1998). In contrast, most of the planted mangrove areas in Gujarat, as described below, have no freshwater source, are sparse and single-species and thus, do not possess many of the above-mentioned features. Single-species mangrove plantations have been argued to provide few ecosystem services, show lower capacity to regenerate, and hence, to be unsustainable in the long run (Rovai et al., 2012). Thus, the gain to the state from this massive investment is questionable; and evaluation, as attempted here, forms an important research issue for sustainability and justification of the resources allocated. Hutchison, Spalding, and zu Ermgassen (2014) provide a comprehensive account of the ecological processes through which mangroves contribute to fishery. ‘‘Mangroves enhance fish production via two main mechanisms—the provision of food and of shelter . . . (p. 6).” Mangroves provide nursery, habitat and nutrients to fish fry and juvenile fishes. Thus, near-coast fisheries (like inshore mixed fisheries, and inshore mollusk and crustacean fisheries) are the most likely and immediate beneficiaries of the habitat services of mangroves. Nonetheless, commercial fisheries that operate many kilometers away from mangroves could also benefit from the nursery habitat role of mangroves and their protection service from predation. Thus, mangroves are likely to be an important determinant of fish stock—the potential fishable biomass of a region, and fish catch—though the sustainability of fishery is influenced more by how it is managed. However, the linkage between mangrove habitat and fishery production is reviewed to be location specific, not universally observed (Saenger, Gartside, & Funge-Smith, 2013).This necessitates a careful examination of the mangrove-fishery linkage for every study area. This study tries to do this for both inshore artisanal and offshore commercial fishery. A set of carefully collected data and panel regression methodology are used for artisanal fishery. Offshore commercial fishery is targeted and vessels are acquired to carry out targeted fishing on pelagic, demersal, mollusks, etc. It is thus natural that vessels acquired, which

are privately owned, will be guided by species availability and previous experience. As advised by fishery ecologists, the presence of mangrove influences the growth and availability of specific fish species like demersal, crustaceans and mollusks, but not pelagic. Whether mangrove presence yields any benefits to the commercial offshore fishery of Gujarat is thus, discussed in terms of these species. The mangrove plantation in Gujarat state is described first which is then followed by a review of studies on mangrove fishery linkage and then the planted mangroves’ contribution to inshore and offshore fishery of Gujarat is evaluated. Both data and the evaluation methodology for these two sectors are different from each other. (a) Mangrove plantation in the state of Gujarat, India In order to measure the extent of planted mangroves, this study assessed the mangrove cover of Gujarat state for three different years: 1939, 1990 and 2013. The source of 1939 data was an open access online source, www.lib.utexas.edu/maps/ ams/india/nf-45-14.jpg whereas Indian Satellite image LANDSAT--TM–1990 and RESOURCE--SAT–2–LISS-III– 2013 with a resolution of 23 m were used to measure the mangrove cover for 1990 and 2013. Table 1 shows the mangrove cover of the coastal districts of Gujarat for these years. Historically, and as can be observed from Table 1, the state had extensive mangrove cover, to the extent of 855 sq km, spread mostly in four districts: Bharuch, Bhavnagar, Jamnagar, and Kutch. By 1990, mangrove cover declined in Bharuch, Bhavnagar, and Jamnagar; whereas they were planted in Ahmedabad, Anand, Kutch, Navsari, Surat, etc., so that the mangrove cover, in 1990, was nearly the same as it was in 1939. By 2013, all coastal districts other than Porbander, Rajkot, and Vadodara had some mangroves. Thus, although natural mangroves have been destroyed in most areas, total mangrove cover has gone up, as plantation is taking place in almost all coastal districts. These estimates also match with the Gujarat Ecology Commission’s estimate of 1,027 sq km of mangroves for 2006 (GEC, 2009). Depending on location, different mangrove plantation techniques have been used in the state (Pandey & Pandey, 2009), like 1. poly plot (PP) plantation in open seashore areas; 2. enrichment plantation (EP) in areas generally having existing sparse natural mangrove vegetation; Table 1. Mangrove cover (in sq km) of Gujarat as assessed from satellite images Districts

Mangrove_ 1939

Mangrove_ 1990

Mangrove_ 2013

Ahmedabad Amreli Anand Bharuch Bhavnagar Jamnagar Junagadh Kachchh Navsari Porbandar Rajkot Surat Vadodara Valsad

3 0 0 81 105 229 0 419 0 2 15 0 0 0

76 0 19 36 19 79 2 604 10 1 1 27 2 1

34 3 9 56 24 300 12 1,198 19 0 0 36 0 4

Total mangrove cover

855

876

1,694

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

ECOLOGICAL RESTORATION AND LIVELIHOOD

3. direct seed sowing (DSS) at blank areas devoid of mangroves; and 4. fish bone channel (FBC) method in areas with poor inundation. On quality of mangroves, the Gujarat mangroves are reported to be sparse; only 16% being moderately dense, and the rest being open mangroves; mostly stunted, with an average height of 1 m (except in South Gujarat); less diverse; and dominated by the Avicenna marina species, especially in the Gulf of Kutch region (north-western part of Gujarat, bordering Pakistan), where it is the only species. This is due to high salinity and lack of fresh water. The mangroves of South Gujarat (Navsari, Surat, and Valsad districts) are more diverse as there are perennial sources of fresh water and there are some 16 different species (DasGupta & Shaw, 2013; Hirway & Goswami, 2007; Pandey & Pandey, 2009; Singh, Ansari, Kumar, & Sarkar, 2012), but South Gujarat holds only three% of the total mangrove cover of the state. Thus, the above-mentioned limiting features apply to the 97% of the mangroves of the state.

3

observe the flow of ecosystem services over time; thus, their assessment is more generic than that of the studies above, which are based on limited samples. One needs to use catch information, along with site features, over a reasonably long period, so that temporal variations are controlled and an unbiased assessment is made of the contribution of planted mangroves to fishery. Studies based on surveys of fishermen report increased fish catch after mangrove plantation in Gujarat (Hirway & Goswami, 2007; Viswanathan, Pathak, & Mehta, 2011), but do not use rigorous evaluation techniques. The creeks in Gujarat have diverse features like having both natural and planted mangroves (called enriched plantations) or having only natural mangroves or only planted mangroves, and have different levels of water pollution from coastal industries, harbors, salt works etc. The research methodology of the study tries to exploit these features to capture the effect of planted mangroves on artisanal coastal fishery. 2. METHODOLOGY

(b) Mangroves and fishery (a) Data and analytical method for inshore fishing Studies report natural mangroves to increase fish growth and contribute to the fishing community’s welfare (Barbier & Strand, 1998; Barbier et al., 2011; Freeman, 1991; Lahmann, Snedaker, & Brown 1987; Ro¨nnba¨ck, 1999; Sathirathai & Barbier, 2001). During 1983–93, in Thailand, when a total area of 300 sq km of mangroves was deforested (at the annual rate of 30 sq km), the coastal fishermen community that practises artisanal as well as offshore fishery was shown to have lost USD408,000 to USD12,000 in annual welfare depending on the elasticity of demand (Barbier, Strand, & Sathirathai, 2002). Studies show the annual value per sq km of mangroves to fisheries to vary from USD0.14 million to USD6.1 million for offshore prawn fisheries, and from just USD34 to USD2.7 million for inshore coastal fisheries (Barbier & Strand, 1998; Christensen, Tarp, & Hjortsø, 2008; Jansen & Padilla, 1999). Studies assessing the contribution of planted mangroves to fishery have mainly followed ecological approaches, like comparing fish abundance in planted mangrove areas to that in natural mangrove areas or in non-mangrove areas, and have come up with diverse results. Some find that planted mangrove areas show similar abundance (or even better) as natural mangrove areas if the planted strand has low elevation and is in good health, but poor abundance if denuded (Crona, Holmgren, & Ro¨nnba¨ck, 2006; Crona & Ro¨nnba¨ck, 2005; Walton, Le Vay, Truong, & Ut, 2006; Walton, SamonteTan, Primavera, Edwards-Jones, & Le Vay, 2006). The first two studies are based in Kenya and the second two in the Philippines. There are studies that show opposite results. A study from Gazi Bay, Kenya, and another from Pasir Ris in the eastern part of Singapore, found that the fish catch from mangrove cleared sites and sandy beaches was more than the catch from reforested mangrove strands (Huxham, Kimani, & Augley, 2004; Jaafar, Hajisamae, Chou, & Yatiman, 2004). However, the authors highlighted the need to control for temporal and site-specific features that affect fish growth and fish catch while evaluating the planted mangroves as nursery ground for fish. A study from Kenya that used the local community’s perception to compare ecosystem services from planted mangroves with that from natural mangroves reported that, in many cases, the benefits from planted strands were nearly one-third of similar services from natural strands (Ro¨nnba¨ck, Crona, & Ingwall, 2007). Local communities

The inshore fishery is mostly a mixed-species fishing performed by local fishermen, called Pagediya in Gujarat. These fishermen are traditional artisanal fishermen who practiced such fishing in nearby creeks. Almost all of them use nets and catch whatever species they get. Creek features with respect to mangrove presence or level of pollution are beyond their control. They move from creek to creek on different days, which mean fish catch on a day is likely to be endogenous to creek selection and data based on a few days’ survey, or on a limited sample survey, can produce biased results. Endogeneity can occur when one of the explanatory variables (viz. the creeks) is a choice variable, not a random variable. In such situation, this variable will be related to the error term and this will result in selection bias or omitted variable bias on the dependant variable. If the fisherman goes to a creek knowingly that he will get high catch there on that day, it means there are some other temporary factors influencing the catch in the creek on that day and one needs to neutralize the effect of these temporary features to get unbiased estimate of the effect of the fixed features of the creek like presence of mangroves. One possible way is to collect catch data for a longer period to neutralize such effects. Thus, the assessment of the effect of planted mangroves on inshore fishery was done 1. in terms of total catch (including all species, not catch by species) per day per fisherman; and 2. by tracking select fishermen for many days, so that there would be data on catch from the same creek by the same fisherman for multiple days. Daily panel data were collected through a survey of Pagediya fishermen. This survey was conducted simultaneously in six different fishermen’s villages of Kutch district of Gujarat in the months of December and January in 2014. 3 In each village 10 fishermen were randomly selected, and the daily catch of each fisherman were recorded on every alternate day for nearly one-and-a-half months. This gave, on average 20 days of information on the catch by each fisherman. 4 This survey was conducted simultaneously in all villages to neutralize the seasonal effect or any day-to-day fluctuation in catch, or in the health of a fisherman, or any outlier catch effect. The daily catches were recorded in the evening after the fishermen returned from sea.

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

4

WORLD DEVELOPMENT

On the first day, the demographic and socio-economic details of each fisherman were recorded, along with fishingrelated details, like, the name of the creek where they had fished, time of departure and arrival, the fishing instrument used, the type of supplementary material that was carried in boat and catch of the day (both the number of species and the total weight). On the subsequent days, only the fishingrelated details were recorded. The same interviewer visited the same fisherman’s house each alternate day to record the information. The villages selected were (with the taluka names in brackets) Vandi (Anjar), Luni and Sekhadiya (Mundra), Modhva (Mandvi), and Budiya, Lala, and Rampar (Abdasa). These villages are spread across 100–120 km of the inner coastline of Kutch district, and face the Gulf of Kutch, which gave enough scope to control for water pollution as well as different types of mangrove habitats. The coastal areas of both Anjar and Mundra are highly industrialized, whereas Mandvi is less industrialized and Abdasa is the least industrialized. Similarly, Anjar has creeks with natural and planted mangroves, Mundra has creeks with planted mangroves or no mangroves at all, and both Mandvi and Abdasa have creeks mostly with open mudflats having no mangroves. This unique survey gave daily information on the fish catch from 14 different creeks of Kutch district that have very different features with respect to levels of pollution, mangrove cover, etc. In total 1029 days of fishing information was collected. Table 7 in the Appendix lists out the catch details per habitat type and the level of pollution. Rather than selecting creeks and taking fish/juveniles samples from there, as has been done in previous studies, (Crona & Ro¨nnba¨ck, 2005; Crona et al., 2006; Walton, Le Vay et al., 2006; Walton, Samonte-Tan et al., 2006), I tracked fishermen, and compared their daily catch from different creeks and measured the planted mangrove effect from these differences indirectly. Univariate tabular comparisons and multivariate panel regression (Eqn. (1)) estimates from the daily catch panel data have been used in the analysis. Panel data have many advantages over time series or cross sectional data as they control for the unobserved group (fishermen here) heterogeneity and produce unbiased results. The panel equation estimated is the following: Y it ¼ a þ bi H i þ cit S it þ uj C j þ V þ B þ eit

ð1Þ

where Y is the catch by the ith fisherman (of village v belonging to block b) on tth day (from creek j). H is vector of household features of ith fisherman, S is vector of items carried in boat by the ith fisherman on the tth fishing day, C is vector of jth creek features, V is village fixed-effect, B is tehsil fixedeffect and e is error. According to fishermen, pollution was the most damaging factor for fishery. The literature also reports that the discharge of heavy metals and effluents, especially oil pollutants, from the large number of industrial units at the coast to have severely affected the ecological health of Gujarat’s mangroves and coast (Jagtap & Nagle, 2007). This study did not have budget to go for detailed pollution test of creek water, hence help of local experts from Gujarat Institute of Desert Ecology (GUIDE), Bhuj was used to rank creeks’ pollution load as either high, or medium, or low (negligible). 5 The study area creeks had pollution coming from ports, thermal power plants and salt pans—salt pans being there all over the coastline. Whereas almost all creeks received some amount of saline brine discharge from salt pans, some received additional pollution from harbors and some from both harbors and power plants. Thus, the location of creeks vis-a-vis pollution sources and their proximity was used in ranking the water pollution

level. Creeks close to harbors and power plants were ranked highly polluted and the ones close to only harbors and far from power plants were ranked moderately polluted. Other information on fishermen’s household and what material they carry when they go for fishing, collected through the survey, were used as additional explanatory variables as these could be affecting their fishing behavior. For example, a more experienced fisherman could be more efficient than a less experienced person and so could be a person who belongs to a fisherman family. Each such factor is explained in the result section. Village and tehsil fixed effects are used to control for any unknown features of a village or any fishermen related policy or union at the tehsil level that could be impacting the artisanal fishermen of that village or tehsil, not others. The possibility of controlling for such unobserved features is one of the best advantages panel data offer. (b) Data and analytical methods for offshore fishery Offshore commercial fishery is open access, and vessels of any state can fish anywhere. If mangroves increase the stock of mangrove-dependent species in the coast of Gujarat, the maximum benefits would go to the vessels of Gujarat, as it is least expensive for them to fish in that region if stock is available. (Of course, the fish could be swimming to other areas and to deep sea, but more fish should be available near mangroves.) Using this assumption, I compare the trend in landing of both mangrove dependent and non-dependent species in Gujarat to that of other neighboring west coast states after controlling for vessels and other factors that could be affecting fish landing. Commercial vessels fish in areas where stocks are available and unload their catch at landing centers depending on the facilities available or prices offered or whichever is convenient to them. So, landing data at a landing center may not reflect the effect of the surrounding ecology. Thus, the effect of mangroves on commercial fishery cannot be captured through an analysis based on landing centers, but rather, through a state-level analysis, as increased catch by any vessel because of more availability would be reported at some landing center and will be reflected in aggregate catch of the state. Hence, to examine the effect of mangroves on commercial fishery, this paper uses state-level landing data by species. Broad species (pelagic, demersal, etc. not individual species) wise data on fish landing for all states were collected from reports of the Central Marine Fishery Research Institute (CMFRI) of India, which collects data by individual species and vessels from major landing centers through an intensive and carefully drafted sampling procedure. Their enumerators collect at least 20 days of sample data each month. Usually, small, far-off landing centers are not represented in CMFRI surveys. However, since most of the landing of a state is reported in major landing centers, and not in small ones, CMFRI catch data are representative of the state as a whole. In contrast, the data collected by the states fishery departments cover all the landing centers of a state, but inter-state comparison using these data is less meaningful as sampling strategy and sample days vary between states. Information on the number of vessels engaged in fishery and on fishery expenditure was collected from fishery surveys of CMFRI and reports of the Department of Animal Husbandry, Dairying and Fisheries of the Ministry of Agriculture, Government of India. Data used are for the period of 1985–2011. First, I use panel regressions to examine the link of mangroves with the types of fish landing (Eqns. (2) and (3)). Then, I measure the effect of planted mangroves on landings of

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

ECOLOGICAL RESTORATION AND LIVELIHOOD

Gujarat using difference-in-differences (DID) regressions (Eqn. (4)). DID technique helps to separate out the effect of an intervention (mangrove plantation in present case) on the treated group with the help of the counterfactual outcome (outcome without intervention) which is derived from a control group that do not receive the intervention, but is otherwise similar to the treatment group. This approach follows a before-treatment, after-treatment, with-treatment and without-treatment set up and the difference in mean of the outcome variable between the treatment and the control group before the treatment is compared to the same difference after the treatment. The difference of these differences is called the treatment effect. To derive DID estimates, the year 1995 is used as the cut-off (pre and post treatment) year and the data of the neighboring coastal states (Maharashtra, Goa, and Karnataka) are used as controls. The choice is for using Maharashtra, Goa, and Karnataka combined as the control group because they are the neighboring states and their fishermen fish in the same Arabian Sea as the fishermen of Gujarat and their combined coastline length of 1,540 km is comparable to the extensive 1,600 km coastline of Gujarat. The year 1995 is chosen as the cut-off year as the mangrove cover of Gujarat is found increasing monotonically only from 1995 (FSI, 2014). This means hectares planted were less in number compared to hectares deforested in the state before this year. So, the aggregative effect of planted mangroves (called treatment effect hence forward) on the fishery sector of Gujarat is measured from 1995 onward. The following three sets of equations have been estimated. Y itj ¼ a0 þ a1 V it þ a2 Eit þ a3 T þ a4 M it þ a5 Rg þ a6 S þ eit

ð2Þ

Y itj ¼ b0 þ b1 V it þ b2 Eit þ b3 T þ b4 DM it þ b5 Rg þ b6 S þ eit ð3Þ Y itj ¼ c0 þ c1 tp þ c2 ts þ c3 tp ts þ c4 V it þ c5 Eit þ c6 T þ c7 Rg þ c8 S þ eit

ð4Þ

where Y is fish landing of ith state (Gujarat, Maharashtra, Karnataka, and Goa) in tth year. Superscript j is fish category (pelagic, demersal, crustaceans, and mollusks). V is fishing vessels, E is fishery expenditure, T is time trend (1985 = 1, 1986 = 2, etc.), M is the mangrove cover, Rg is fishery policy regime shift dummy for Gujarat (=0 if year is <1989, and =1 for other years), S is state fixed-effect, DM is change in mangrove cover, and e is error term. The result section explains the need for using a regime shift dummy for Gujarat. State fixed effects control for the fixed or time non-varying features of a state like coastline length, traditional knowledge of fishermen, etc. There may be features other than the ones used in the regressions affecting commercial fishery or being more representative of the fishing effort (like vessel size or horse power), but it was difficult to arrange such state level information. However, state fixed effects control for some of these interstate variations. In Eqn. (4), tp is treatment period dummy (=0 before 1995 and =1 after 1995), ts is treatment state dummy (=1 for Gujarat and =0 for other states). If mangrove plantation affects any type of catch, that effect will be captured by a significant value of c3 the coefficient of the interaction term of treatment period and treatment state. This term captures the difference in average catch of Gujarat over the other three west coast states in post-1995 period compared to such difference in their average catch in the pre-1995 period. This difference, if significant, is assumed to be attributable to mangrove plantation in Gujarat as other likely important factors

5

(fishing vessels, fishery expenditure, changes in fishery policies, fixed features of states) affecting catch is accounted for in the model. To check presence of confounders that could have caused such results, the results on c3 are counterchecked by altering the treatment state, i.e., taking one of the other states as treatment and rest as control and re-estimating Eqn. (4), as explained later. The statistical package STATA is used to derive all results. 3. RESULTS (a) Results on inshore fishery Data from fishermen survey show that most of the Pagediya fishermen (80%) were illiterate, and had 16 years of fishing experience on average. Very few (20%) had a subsidiary occupation, and their average monthly household expenditure was roughly INR9,000 (USD135). Fishermen of each village spent 9–10 h on fishing per fishing day. They were unanimous on two things: (1) water pollution has decreased fish catch and (2) mangroves, wherever they were being planted, are not helping fishery. Next, the collected data were examined statistically. First, creeks were grouped on the basis of presence of mangroves. Table 2 shows the average daily catch from these creeks and compares the mean catch. Column 4 shows the number of species caught daily and the highest numbers, 3.2 species on average, are caught from creeks with no mangroves. The number of species caught is 2.88 from creeks having natural mangroves, 3 from creeks having enriched plantation, 6 and the lowest 2.5, is from creeks having planted mangroves. A two-sample t test of the difference between the mean catch from no mangrove creeks vs. the mean catch from other type of creeks is found significant as the t values reported in column 5 proves. Thus, the number of species available seems to be lower in creeks with natural mangroves and the lowest in planted mangroves compared to creeks with no mangroves. However, when the average weight of the species caught is compared, as shown in column 6, the conclusion changes. The weight of the catch is the highest, 6.97 kg, from creeks with enriched plantation followed by that from creeks with natural mangroves, 6.16 kg, and then from creeks with no mangroves (4.81 kg). The weight of catch is least in creeks with planted mangroves (3.69 kg). The differences between the catch from no mangrove creek vs. other creeks are also statistically significant as the t values reported in column 7 proves. Natural mangrove areas seem to have the healthiest fishes and yield the highest catch. Creeks with plain mud flats that have no mangroves have a high density of fish species, but they seem to be unhealthy and low in weight (see Table 7 for more details). Creeks with planted mangroves had the lowest catch in terms of number of species, and also have the lowest catch in terms of weight; this indicates that mangrove plantations do not contribute to fishery in increasing either density or growth, at least in the short run. Mangroves were planted in these creeks during 2006–09, so five-to-eight-year-old mangroves do not seem to help fishery. Fishermen too had the same report. Next, these results were re-examined in panel multivariate regression after using control variables for pollution and fishermen specific features. Table 3 shows these results. In Table 3, the dependent variable is the daily fish catch (weight) of the fishermen interviewed which is explained by a number of variables like creek features (type of mangrove present, water pollution level), education, and experience of

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

Mean (I) > mean (IV) t = 9.22, P = 0.00 3.69 (0.87) Mean (I) > mean (IV), t = 14.56, P = 0.00 IV

Note: Figures in parentheses in column 4 and 6 are standard deviations. All means were statistically different from zero.

3 (0.00)

Creeks having enriched plantations (mangrove plantation in places where sparse natural mangrove existed) Creeks with few year old planted mangroves III

16

2.5 (0.50)

6.97 (1.67)

mean (II), = 0.00 mean (III), = 0.08 > P > P 3.2 (0.63) 2.88 (0.70) 690 133 Creeks with no mangrove Creeks with natural mangroves I II

190

< P < P 4.81 (1.62) 6.16 (1.86)

– Mean (I) t = 5.62, Mean (I) t = 1.40,

– Mean (I) t = 8.53, Mean (I) t = 5.25,

Average weight of species caught t-Value with associated probability for difference in mean of number of species caught between creek type I and other types Average number of species caught Number of fishing days Description of creeks Type of creeks

Table 2. Average daily fish catch of Pagediya fishermen from different types of creeks

mean (II), = 0.00 mean (III) = 0.00

WORLD DEVELOPMENT

t-Value with associated probability for difference in mean of weight of the species caught between creek type I and other types

6

fishermen, family features, what they carry in boat when they go for fishing, etc. As explained above, water pollution was categorized as low, medium, and high depending on the number of sources of pollution near the creeks and their distances from the creeks. The creeks of Abdasa Taluka have low pollution whereas the creeks of other talukas have medium or high pollution depending on their locations. Three mangrove dummy variables corresponding to creeks with natural mangroves, enriched plantations, and planted mangroves are used. The catch from these creeks are compared with the catch from creeks with no mangroves. I wanted to check if people carry fish feed or preservatives or any other material that increased their catch or preserve the catch better that fetched a higher price in the market, but fishermen came with answers like bike, blanket, rope, etc. The results show the items carried while going for fishing or the household specific features to have insignificant effect on catch except the variable ‘number of children’ which depicts a negative and significant effect. The results on the type of habitats show fishermen to be catching 4.237 kg more in natural mangrove creeks, 3.962 kg higher in enriched plantation creeks and 0.948 kg higher in planted mangroves creeks compared to the catch from no mangrove (or plain mud flat) creeks if the pollution level of creeks and the other demographic and socio-economic features of the fishermen are controlled. All these results are highly significant. Similarly, when compared to low pollution creeks, the daily catch is 3.008 kg lower in medium polluted creeks and 4.069 kg lower in highly polluted creeks. (b) Results on offshore commercial fishery The state of Gujarat is the leading marine fishery state on the west coast of India (Figure 1). In 2013, Gujarat contributed 0.72 million tons (19%) out of the 3.78 million tons of fish landing of the country (CMFRI, 2014). Of the four states on the west coast (Gujarat, Maharashtra, Goa, and Karnataka), the state of Maharashtra was the leading state in fish landing until about 1990. Gujarat took the lead that year, and retained the position until 2013 (Figure 1). It is to be noted that the state witnessed the least landing in 1988. There seems to have been a regime shift in the marine sector of the state afterward, and landings increasing steadily. The state of Karnataka has also witnessed an increasing trend in marine catch in the years after 2004–05. Though the state fishery policy is the most important reason for the turnaround of the fishery sector, it needs to be examined if mangroves played any role? As per the Forest Survey of India (2014) report, the mangrove cover in Gujarat was 1,103 sq km in 2013 compared to 186 sq km in Maharashtra, 3 sq km in Karnataka and 22 sq km in Goa (FSI, 2014). As Gujarat has a sizeable mangrove cover compared to others, the increased catch in Gujarat should ideally be in mangrove dependant fish species and in other states, in non-dependant species. State fixed-effect estimates were calculated for Eqns. (2) and (3) by taking total landing and each of its components (pelagic, demersal, crustaceans, and mollusks) as dependent variables. Table 4 shows only the estimated coefficients of mangrove cover from Eqn. (2) and change in mangrove cover from Eqn. (3) (detailed results available on request). The state’s mangrove cover has a significant and positive link with total, demersal, crustaceans, and mollusks landing, but not with pelagic landing. This is consistent with the general observations from other areas that pelagic species are not mangrove-dependent, whereas, the other three categories are. Replacing total mangrove area by change in mangrove

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

ECOLOGICAL RESTORATION AND LIVELIHOOD

7

Table 3. Fixed-effect estimates for inshore mixed fishery in Kutch district based on Pagediya fishermen’s survey (Dependent variable = Weight of the daily catch) Independent variables

Estimated coefficient

Natural mangrove creek Enriched plantation creek Planted mangrove creek Medium polluted creek Highly polluted creek Fishing experience (years) Has other subsidiary occupation Ancestors were fishermen Whether educated Number of male members Number of female members Number of children Carry bike in boat Carry blanket in boat Carry food stock in boat Carry ropes in boat Constant Joint significance test of mangrove coefficients being different than zero

4.237*** (7.96) 3.932*** (7.24) 0.948*** (2.71) 3.008*** (8.85) 4.069*** (7.25) 0.009 (1.34) 0.143 (1.42) 0.353 (0.88) 0.180 (1.77) 0.079 (1.57) 0.025 (0.48) 0.070** (2.16) 0.304 (0.74) 0.164 (0.36) 0.153 (0.90) 0.129 (0.54) 6.087*** (8.85) Chi2(2) = 1322.89***

Number of observations = 1,029 Number of groups = 57 Observations per group = 15–20 Wald chi2 (16) = 1950.69 (P = 0.00) Note: Figures in parenthesis are absolute t-values.

*** **

,

imply 1%, and 5% level of significance respectively.

Figure 1. Total Marine Catch (in tons) of West Coast States of India.

Table 4. Coefficient estimates of commercial fish catch and mangrove cover in west coast of India (period of analysis: 1985–2011; state fixed-effect estimates) Variables

Dependent variables (in ‘000) Total catch

Total mangrove cover (Eqn. (2)) Change in mangrove cover (Eqn. (3))

Pelagic catch

***

0.285 (4.10) 0.543** (2.13)

Note: Figures in parenthesis are absolute t values and

*** ** *

,

,

0.029 (0.82) 0.165 (1.35)

Demersal catch ***

0.124 (4.51) 0.217** (2.15)

Crustaceans catch

Mollusks catch

***

0.0302*** (4.36) 0.023 (0.88)

0.1025 (5.30) 0.138* (1.85)

imply 1%, 5% and 10% level of significance respectively.

area in the regression (Eqn. (3)) also showed a similar relation, though the level of significance is found low. In both cases, pelagic catch is insignificantly linked to the mangrove area or change in mangrove area of the states. Next, DID estimates are measured from Eqn. (4). Random effect estimates are used to show the coefficients of time invari-

ant features like the state dummies. Table 5 shows the estimates taking Gujarat state as the treatment state and the other west coast states as the control. As observed from the coefficients of treatment effect (which is assumed to capture the planted mangrove effect), all types of catch other than of pelagic species have significantly increased in Gujarat than

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

8

WORLD DEVELOPMENT Table 5. Random effect coefficient estimates of different species of commercial fishcatch of Gujarat compared to other west coast statesa Explanatory variables

Dependent variables (catch in ‘000tons) Total catch

Pelagic catch

Demersal catch

Crustaceans catch

Mollusks catch

Treatment period (high 14.554 (0.55) 12.431 (0.95) 1.219 (0.12) 1.996 (0.27) 1.345 (0.50) mangrove plantation period: 1995–2011) Treatment state (Gujarat) 202.815*** (4.53) 89.283*** (3.99) 93.923*** (5.25) 19.161 (1.51) 0.449 (0.1) 30.163 (1.58) 50.821*** (3.33) 44.717*** (4.12) 11.481*** (2.93) Treatment effect (treatment 137.182*** (3.58) pd  treatment state) Time_trend 1.51 (0.85) 0.876 (0.99) 0.236 (0.33) 0.107 (0.21) 0.504*** (2.77) Fishing vessels 0.001 (0.67) 0.001 (1.46) 0.0003 (0.58) 0.00003 (0.09) 0.0004** (2.57) 0.004** (2.30) 0.002 (1.43) 0.0005 (0.54) 0.0001 (0.43) Fishery expenditure 0.005* (1.61) 1989 onward dummy for 156.540*** (3.88) 58.109*** (2.89) 44.011*** (2.74) 46.935*** (4.11) 7.485* (1.81) Gujarat Karnataka state dummy 137.318*** (6.53) 77.205*** (7.35) 36.18*** (4.31) 19.658*** (3.30) 4.274** (1.98) *** *** *** *** 79.240 (7.78) 77.501 (9.51) 95.161 (16.44) 15.614*** (7.47) Maharashtra state dummy 267.516 (13.10) Constant 61.453*** (3.30) 48.463*** (5.21) 5.437 (0.73) 13.097** (2.48) 5.545*** (2.9) Wald chi sq value Wald chi2(9) = 791.83 Wald chi2(9) = 332.77 Wald chi2(9) = 730.26 Wald chi2(9) = 757.13 Wald chi2(9) = 379.9 Number of observations 108 108 108 108 108 Note: Figures in parenthesis are absolute t values and a Fixed effect estimates also give similar results.

*** ** *

,

,

imply level of significance to be 1%, 5%, and 10% respectively.

Table 6. Random effect coefficients of interaction between treatment period (1995–2011) and one of west coast state Treatment effect for West coast states

Total catch

   

***

Treatment Treatment Treatment Treatment

Period Period Period Period

Gujarat Maharashtra Karnataka Goa

Pelagic catch

137.18 (3.58) 26.64 (0.78) 14.23 (0.98) 69.36** (2.36)

Note: Figures in parenthesis are absolute t values and

*** ** *

,

,

30.16 (1.58) 18.49 (1.14) 5.98 (0.42) 8.44 (0.59)

Demersal catch ***

50.82 (3.33) 24.62* (1.84) 13.11 (1.12) 22.36* (1.91)

Crustaceans catch ***

44.72 (4.12) 12.81 (1.30) 7.90 (0.92) 26.33*** (3.17)

Molluscs catch 11.48*** (2.93) 3.66 (1.07) 3.04 (1.02) 12.25*** (4.43)

imply level of significance to be 1%, 5%, and 10% respectively.

in other states in the post-1995 period compared to the pre1995 period after other variables (like vessels, fishery expenditure by governments, state-specific fixed features, etc.) are controlled. 7 The pelagic catch in Gujarat shows no significant increase during 1995–2011 compared to other west coast states. The coefficients of treatment period dummy variable are mostly negative and insignificant which means the catch in the west coast of India has been similar to what it was during 1985– 94, but Gujarat has experienced significantly higher catch than other states. To cross-check for presence of any other confounding factors not controlled in the model and responsible for the increased catch in Gujarat in post-1995 period, the same set of equations was estimated repeatedly taking each of the other states as treatment state. If either Maharashtra or Karnataka or Goa shows increased catch for any of the mangrove-dependent fish categories during 1995–2011, then the hypothesis that the increased catch of Gujarat during this period is due to mangrove plantations, will be in doubt. Therefore, Eqn. (4) was re-estimated repeatedly after changing the treatment state from Gujarat to one of the other states. Table 6 shows just the estimated coefficients of the interaction terms for different species of fish; the results for Gujarat as treatment state (second row) are same as shown in Table 5. Table 6 shows that Gujarat was the only state to have attained significant increase in catch of mangrove dependant species during the post-1995 period. This finding is consistent with the positive effect of mangroves to fish catch in general and adds credence to the assertion that mangrove-plantation has contributed toward increased catch in Gujarat.

4. VALUING THE CONTRIBUTION OF PLANTED MANGROVES TO THE FISHERY SECTOR OF GUJARAT Due to increased mangrove cover since 1995, the commercial fishery sector of the state has been witnessing an increased annual catch of 50,821 tons of demersal species, 44,717 tons of crustaceans, and 11,481 tons of mollusks (Table 5) which is approximately 15% of the total annual landing of the state. Pelagic catch is excluded, as such species are less dependent on mangroves, and also because the plantation effect on pelagic catch is not significant in Table 5. Using average approximate local market prices of INR250 per kg for demersal, INR500 per kg for crustaceans and INR80 per kg of mollusks, the annual monetary gain to the commercial fishery sector of Gujarat due to planted mangroves is measured to be INR36 billion (USD0.57 billion) at 2013–14 prices. This is the annual gain to all the commercial offshore fishermen engaged in demersal, crustacean, and mollusk fishery. The effect of planted mangroves on inshore coastal fishery, described in Table 3, shows that mangroves planted in mud flats increase fish catch by 0.948 kg more per day per fisherman compared to no-mangrove areas in the villages studied. The survey data also showed that mangroves planted in sparse natural mangrove areas, or enriched mangrove plantations, increased fish catch by 3.923 kg more per day, and that natural mangroves increased it by 4.237 kg per day compared to no mangrove area. However, these increases are not easily detected being nullified by the decline in catch due to pollution from ports and industries. Detailed data on year-wise and

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

ECOLOGICAL RESTORATION AND LIVELIHOOD

location-wise mangrove plantation and survival being unavailable, it was difficult to do a systematic valuation of mangroves’ contribution to inshore fishing. Taking the above values as marginal effects of the habitats, the average contributions of mud flat planted and enriched plantation mangroves (2.436 kg per day per fisherman) is used to calculate the value of planted mangroves to inshore fisheries. There are around 8,000 Pagediya fishermen families in Gujarat and they fish approximately 15 days a month (mostly follow tidal cycle) throughout the year. 8 Assuming that they all are being similarly benefited due to mangrove plantation, the yearly contribution of planted mangroves to inshore fishing comes to INR0.035 billion (USD5.6 million), using local market price of INR100 per kg as the price of mixed fishes caught by Pagediya fishermen in Gujarat. Thus, the contribution of planted mangroves to both inshore and offshore fisheries of Gujarat calculates to INR36.04 billion (USD0.57 billion) per year or INR0.44 million (USD7002) per hectare per year at 2013–14 prices as there are approximately 817 sq km of planted mangroves in the state during 1995–2011. 5. DISCUSSION AND CONCLUSION The state of Gujarat has achieved remarkable success in planting thousands of hectares of mangroves over coastal mud flats. This study used open access online source, www. lib.utexas.edu/maps/ams/india/nf-45-14.jpg, and satellite images of Indian Satellite LANDSAT--TM–1990 and RESO URCE--SAT–2–LISS--III–2013 to assess and measure the mangrove cover in the state in 1939, 1990, and 2013. The mangrove cover of the state has nearly doubled from 855 sq km in 1939 and 876 sq km in 1990 to 1694 sq km in 2013. The study assessed the contribution of these planted mangroves to the economy of Gujarat by examining the role of mangroves as nursery ground and habitat for fish fry. A survey of artisanal fishermen that maintained data on the daily catch of 57 fishermen over a period of one-and-a-half months was used to estimate the contribution of planted mangroves to inshore artisanal fishery. To measure the contribution of planted mangroves to offshore commercial fishery, the study used catch data by species from 1985 to 2011 from the reports of the Central Marine Fishery Institute of India. Panel regression and difference-in-differences techniques were used to estimate the models and calculate the enhanced catch attributable to planted mangroves. The results support that mangrove plantations have positive effect on fisheries. Fishermen who fish in creeks where mangroves were recently planted (within the past five to eight years) catch approximately 0.948 kg more daily on an average compared to fishermen who fish in creeks with no mangroves, though the catch is much higher in creeks with natural mangroves and with enriched plantations. However, this result is detectable only after one controls for the pollution level in the creeks. Comparing the increase in catches, the contribution of planted mangroves to the fishery is found to be only 22% of that of the natural mangroves of the area, which is similar to the 25% relative flow of ecosystem services from ecological restoration sites globally compared to natural sites (Benayas, Newton, Diaz, & Bullock, 2009). The planted mangrove effect on commercial fishery of Gujarat is measured for post-1995 period when mangrove cover has sharply increased in the state. Difference-inDifferences technique is used with the neighboring west coast

9

states of Gujarat (Maharashtra, Goa, and Karnataka combined) as a control group to measure the counterfactual outcome. Analysis shows the catch of mangrove dependant species like demersals, mollusks, and crustaceans, but not the catch of pelagic species to have significantly increased in Gujarat in post-1995 period, and these increases, approximating 15% of the total annual landing of the state, are attributed to mangrove plantation. Offshore fishery being open access, there are chances that the benefit of increased stock due to increase in mangrove cover in Gujarat may accrue to vessels of neighboring states as well. However, there are political, locational and economic reasons to believe that maximum benefit will be reaped by Gujarat vessels. Gujarat forms the north-western border between India and Pakistan and there are political reasons for not allowing cross-border fishing. The chances of Pakistani vessels fishing along the Gujarat coast are very low and so are the chances of Gujarat vessels fishing along the Pakistan coast. For Gujarat is the northernmost state, it is also expensive for the other west coast state vessels to travel so far and fish near Gujarat coast. So, the increased fish stock due to increased mangrove cover in Gujarat is likely to be caught mostly by Gujarat vessels. When compared with Maharashtra, Goa, and Karnataka, the average landing center prices of important fish species are the lowest in Gujarat (CMFRI, 2014). This also rules out the vessels of other states to be reporting their catch in Gujarat; rather, there are chances that vessels of Gujarat could be reporting their catch in these states to take advantage of high prices. Thus, the catch data used in the analysis may be underestimating the actual catch of Gujarat and, to that extent, the results of Table 5 may be underestimating the planted mangrove contribution to Gujarat fishery, but chances of overestimation, or wrong attribution, are low. Of course, there could be other natural or anthropogenic confounders like change in rainfall pattern leading to change in nutrient load in water or change in sewage disposal system of states, etc. which could be causing change in catch rates and the results are conditional to such confounders not being present. The annual contribution of all planted mangroves was calculated to be INR36 billion to offshore commercial fishery and INR0.035 billion to inshore coastal fishery which add to a total annual contribution of INR36.04 billion to the fishery sector of the state. This calculates to a unitary contribution of INR0.44 million (USD7002) per hectare per year. This is a significant contribution, despite many limiting features of planted mangroves in Gujarat (like single-species, being over mud flats with high salinity, being stunted, and with limited, or no freshwater sources, etc.). Though some mollusk species are considered mangrove non-dependant, mollusks are included in valuation as it was difficult to separate out the individual species from the total mollusks catch data provided by CMFRI. As mollusks had a very low coefficient for post-1995 period in Gujarat and also a very low value, excluding these species will marginally change the value of planted mangroves to fishery. A cost-benefit analysis could not be carried out as detailed information on plantation, survival and cost involved was not available. Nonetheless, restoration/plantation seems to be a welcome attempt toward biodiversity enhancement and conservation, and benefits will accrue if mangroves are preserved. With mangrove plantation happening at many places, this study makes an important contribution to the literature by assessing the contribution of planted mangroves to the economy of Gujarat state.

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

10

WORLD DEVELOPMENT

NOTES 1. http://gfw.blog.s3.amazonaws.com/2015/02/Tree-Cover-Loss-charta_global.jpg, accessed on 4th April 2015.

5. Marine ecologist, Prof A. Thivakaran ([email protected]) of GUIDE kindly helped in this ranking.

2. Convention on Biological Diversity, The Ecosystem Approach (UNEP/ CBD/COP/5/23 Decision V/6, Nairobi, Kenya 2000).

6. Enriched plantation is plantation in sparse natural mangrove areas.

3. Kutch district has largest mangroves (Table 1) and the mangroves here have all the limiting features described in the text. So the results are quite representative of the values of mangroves of Gujarat.

7. Interestingly, treatment effect becomes insignificant if mangrove area is included in this regression (results available on request). 8. Data received from Bharat Patel (+91 9426469803), leader of a grass root Pagediya Fishermen association.

4. The survey could be completed only for 57 fishermen.

REFERENCES Aburto-Oropeza, O., Ezcurra, E., Danemann, G., Valdez, V., Murray, J., & Sala, E. (2008). Mangroves in the Gulf of California increase fishery yields. Proceedings of the National Academy of Sciences, 105(30), 10456–10459. Barbier, E. B., Hacker, S. D., Kennedy, C., Koch, E. W., Stier, A. C., & Silliman, B. R. (2011). The value of estuarine and coastal ecosystem services. Ecological Monographs, 81(2), 169–193. Barbier, E. B., Koch, E. W., Silliman, B. R., Hacker, S. D., Wolanski, E., Primavera, J., ... Reed, D. J. (2008). Coastal ecosystem-based management with nonlinear ecological functions and values. Science, 319(5861), 321–323. Barbier, E. B., & Strand, I. (1998). Valuing mangrove-fishery linkages—A case study of Campeche, Mexico. Environmental and Resource Economics, 12(2), 151–166. Barbier, E. B., Strand, I., & Sathirathai, S. (2002). Do open access conditions affect the valuation of an externality? Estimating the welfare effects of mangrove-fishery linkages in Thailand. Environmental and Resource Economics, 21(4), 343–365. Benayas, J. M. R., Newton, A. C., Diaz, A., & Bullock, J. M. (2009). Enhancement of biodiversity and ecosystem services by ecological restoration: a meta analysis. Science, 325(5944), 1121–1124. Blaber, S. J. (2007). Mangroves and fishes: issues of diversity, dependence, and dogma. Bulletin of Marine Science, 80(3), 457–472. Bosire, J. O., Dahdouh-Guebas, F., Walton, M., Crona, B. I., Lewis, R. R., Field, C., ... Koedam, N. (2008). Functionality of restored mangroves: A review. Aquatic Botany, 89(2), 251–259. Christensen, S. M., Tarp, P., & Hjortsø, C. N. (2008). Mangrove forest management planning in coastal buffer and conservation zones, Vietnam: A multi-methodological approach incorporating multiple stakeholders. Ocean & Coastal Management, 51(10), 712–726. CMFRI (Central Marine Fishery Research Institute) (2014). CMFRI 2013-14, annual report. Kochi, India: CMFRI, Indian Council of Agricultural Research. Crona, B. I., Holmgren, S., & Ro¨nnba¨ck, P. (2006). Re-establishment of epibiotic communities in reforested mangroves of Gazi Bay, Kenya. Wetlands Ecology and Management, 14(6), 527–538. Crona, B. I., & Ro¨nnba¨ck, P. (2005). Use of replanted mangroves as nursery grounds by shrimp communities in Gazi Bay, Kenya. Estuarine, Coastal and Shelf Science, 65(3), 535–544. Das, S., & Cre´pin, A. S. (2013). Mangroves can provide protection against wind damage during storms. Estuarine, Coastal and Shelf Science, 134, 98–107. Das, S., & Vincent, J. R. (2009). Mangroves protected villages and reduced death toll during Indian super cyclone. Proceedings of the National Academy of Sciences, 106(18), 7357–7360. DasGupta, R., & Shaw, R. (2013). Changing perspectives of mangrove management in India—An analytical overview. Ocean & Coastal Management, 80, 107–118. Ellison, A. M. (2008). Managing mangroves with benthic biodiversity in mind: moving beyond roving banditry. Journal of Sea Research, 59(1), 2–15. Erftemeijer, P. L., & Lewis, R. R. (1999). Planting mangroves on intertidal mudflats: habitat restoration or habitat conversion. In: Proceedings of

the ECOTONE VIII seminar enhancing coastal ecosystems restoration for the 21st century, Ranong, Thailand (pp. 23–28). FAO (Food and Agriculture Organization of the United Nations) (2008). The world’s mangroves 1980-2005, FAO Forestry Paper, 153. Rome: FAO. Field, C. D. (1999). Rehabilitation of mangrove ecosystems: An overview. Marine Pollution Bulletin, 37(8), 383–392. Forest Survey of India (2014). India state of forest report 2011. Dehradun: Forest Survey of India, Online at: http://fsi.nic.in/cover_2013/mangroves_in_india.pdf. Freeman, A. M. (1991). Valuing environmental resources under alternative management regimes. Ecological Economics, 3(3), 247–256. Gazetteers (Gujarat State Gazetteers: Kutch District), (1971). Directorate of Government Print, Stationery and Publications, Gujarat State, Ahmedabad. GEC (Gujarat Ecology Commission) (2009). Mangrove’s atlas of Gujarat state. Gandhinagar: Gujarat Ecology Commission, 73p.. Hirway, I., & Goswami, S. (2007). Valuation of coastland resources: The case of mangroves in Gujarat. Ahmedabad: Academic Foundation. Hutchison, J., Spalding, M., & zu Ermgassen, P. (2014). The role of mangroves in fisheries enhancement. The Nature Conservancy and Wetlands International, 54 pages. Huxham, M., Kimani, E., & Augley, J. (2004). Mangrove fish: A comparison of community structure between forested and cleared habitats. Estuarine, Coastal and Shelf Science, 60(4), 637–647. Jaafar, Z., Hajisamae, S., Chou, L. M., & Yatiman, Y. (2004). Community structure of coastal fishes in relation to heavily impacted human modified habitats. Hydrobiologia, 511, 113–123. Jagtap, T. G., & Nagle, V. L. (2007). Response and adaptability of mangrove habitats from the Indian subcontinent to changing climate. Ambio, 36(4), 328–334. Jansen, R., & Padilla, J. E. (1999). Preservation or conversion? Valuation and evaluation of a mangrove forest in the Philippines. Environmental and Resource Economics, 14(3), 297–331. Kairo, J. G., Lang’at, J. K., Dahdouh-Guebas, F., Bosire, J., & Karachi, M. (2008). Structural development and productivity of replanted mangrove plantations in Kenya. Forest Ecology and Management, 255 (7), 2670–2677. Kaly, U. L., & Jones, G. P. (1998). Mangrove restoration: A potential tool for coastal management in tropical developing countries. Ambio, 27(8), 656–661. Khayat, J. A., & Jones, D. A. (1999). A comparison of the macrofauna of natural and replanted mangroves in Qatar. Estuarine Coastal and Shelf Science, 49, 55–63. http://dx.doi.org/10.1016/S0272-7714(99)80009-2. Lahmann, E. J., Snedaker, S. C., & Brown, M. S. (1987). Structural comparisons of mangrove forests near shrimp ponds in southern Ecuador. Interciencia, 12, 240–243. Lewis, R. R. (2001). Mangrove restoration-Costs and benefits of successful ecological restoration. In Proceedings of the Mangrove Valuation Workshop. Penang: Universiti Sains Malaysia (Vol. 4, No. 8). Lewis, R. R. (2010). Mangrove field or dreams: if we build it, will they come? SWS research Brief No. 2009-005. 4 p. Wetland Science and Practice, 27(1), 15–18.

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010

ECOLOGICAL RESTORATION AND LIVELIHOOD Lewis, R. R. (2009). Methods and criteria for successful mangrove forest restoration. In G. M. E. Perillo, E. Wolanski, D. R. Cahoon, & M. M. Brinson (Eds.), Coastal Wetlands: An Integrated Ecosystem Approach (pp. 787–800). Amsterdam: Elsevier Press. Macintosh, D. J., Ashton, E. C., & Havanon, S. (2002). Mangrove rehabilitation and intertidal biodiversity: A study in the Ranong mangrove ecosystem, Thailand. Estuarine, Coastal and Shelf Science, 55(3), 331–345. McKee, K. L., & Faulkner, P. L. (2000). Restoration of biogeochemical function in mangrove forests. Restoration Ecology, 8(3), 247–259. MEA (Millennium Ecosystem Assessment) (2005). Ecosystems and human well-being: Synthesis. Washington, DC: Island. Meyfroidt, P., & Lambin, E. F. (2009). Forest transition in Vietnam and displacement of deforestation abroad. Proceedings of the National Academy of Sciences, 106(38), 16139–16144. Mukherjee, N., Sutherland, W. J., Dicks, L., Huge´, J., Koedam, N., & Dahdouh-Guebas, F. (2014). Ecosystem service valuations of mangrove ecosystems to inform decision making and future valuation exercises. PLoS One, 9(9). http://dx.doi.org/10.1371/journal. pone.0107706. Naber, H., Lange, G.M., & Hatziolos, M. (With inputs from The One Ocean Programme, UNEP/WCMC), (2008). Valuation of marine ecosystem services: A gap analysis. Available online at: https://www. cbd.int/marine/voluntary-reports/vr-mc-wb-en.pdf, Accessed on 25th January 2017. Pandey, C. N., & Pandey, R. (2009). The status of mangroves in Gujarat. In Towards Conservation and Management of Indian Mangroves. New Delhi: Ó MOEF, Government of India. Pandey, C. N., Pandey, R., & Khokhariya, B. (2012). Potential area mapping for mangrove restoration in South Gujarat. Gandhinagar: Gujarat Ecological Education and Research (GEER) Foundation, 248 pp.. Ro¨nnba¨ck, P. (1999). The ecological basis for economic value of seafood production supported by mangrove ecosystems. Ecological Economics, 29(2), 235–252. Ro¨nnba¨ck, P., Crona, B., & Ingwall, L. (2007). The return of ecosystem goods and services in replanted mangrove forests: perspectives from local communities in Kenya. Environmental Conservation, 34(04), 313–324. Rovai, A. S., Soriano-Sierra, E. J., Pagliosa, P. R., Cintro´n, G., SchaefferNovelli, Y., Menghini, R. P., ... Dutra, S. J. (2012). Secondary succession impairment in restored mangroves. Wetlands Ecology and Management, 20(5), 447–459.

11

Saenger, P., Gartside, D., & Funge-Smith, S. (2013). A review of mangroves and seagrass ecosystems and their linkage to fisheries and fisheries management. Bangkok, Thailand: FAO Regional Office for Asia and the Pacific, RAP Publication 2013/09, 74 pp. Sathirathai, S., & Barbier, E. B. (2001). Valuing mangrove conservation in southern Thailand. Contemporary Economic Policy, 19(2), 109–122. Singh, A. K., Ansari, A., Kumar, D., & Sarkar, U. K. (2012). Status, biodiversity and distribution of mangroves in India: An overview. Marine Biodiversity, Uttar Pradesh State Biodiversity Board, pp. 59–67. Valiela, I., Bowen, J. L., & York, J. K. (2001). Mangrove forests: One of the world’s threatened major tropical environments. Bioscience, 51(10), 807–815. Viswanathan, P. K., Pathak, K. D., & Mehta, I. (2011). Socio-economic and ecological benefits of mangrove plantation: A study of community based mangrove restoration activities in Gujarat. Report of sponsored study No. id: 3852, Gujarat Ecology Commission, Gandhinagar, available at: http://www.geciczmp.com/data/sites/1/docs/socio-economic-benefits-of-mangroves.pdf. Walters, B. B. (2000). Local mangrove planting in the Philippines: Are fisher folk and fishpond owners effective restorationists?. Restoration Ecology, 8(3), 237–246. Walton, M. E., Le Vay, L., Truong, L. M., & Ut, V. N. (2006). Significance of mangrove–mudflat boundaries as nursery grounds for the mud crab, Scylla paramamosain. Marine Biology, 149(5), 1199–1207. Walton, M. E., LeVay, L., Lebata, J. H., Binas, J., & Primavera, J. H. (2007). Assessment of the effectiveness of mangrove rehabilitation using exploited and non-exploited indicator species. Biological Conservation, 138(1), 180. Walton, M. E., Samonte-Tan, G. P., Primavera, J. H., Edwards-Jones, G., & Le Vay, L. (2006). Are mangroves worth replanting? The direct economic benefits of a community-based reforestation project. Environmental Conservation, 33(04), 335–343.

APPENDIX A.

Table 7. Details of catch with detailed creek features Mangrove type in creek

Pollution level in creek

Number of creeks

Number of fishing days

Mean weight of catch (Std. Dev)

Mean number of catch (Std. Dev)

Natural Natural Natural Enriched plantation Enriched plantation Enriched plantation Planted in mudflat Planted in mudflat Planted in mudflat No mangrove No mangrove No mangrove

Low Medium High Low Medium High Low Medium High Low Medium High

– 2 5 – 1 – – 4 – 1 1 –

– 13 120 – 16 – – 190 – 540 150 –

– 7.02 (1.59) 6.06 (1.87) – 6.97 (1.67) – – 3.69 (0.87) – 5.46 (0.83) 2.48 (1.65) –

– 3.15 (0.56) 2.85 (0.71) – 3.0 (0) – – 2.5 (0.50) – 3.20 (0.49) 3.26 (0.97) –

Available online at www.sciencedirect.com

ScienceDirect

Please cite this article in press as: Das, S. Ecological Restoration and Livelihood: Contribution of Planted Mangroves as Nursery and Habitat for Artisanal and Commercial Fishery, World Development (2017), http://dx.doi.org/10.1016/j.worlddev.2017.02.010