Artisanal Fisheries on Kenya's Coral Reefs

Fisheries Research 186 (2017) 177–191

Contents lists available at ScienceDirect

Fisheries Research

journal homepage: www.elsevier.com/locate/fishres

Full length article

Artisanal ﬁsheries on Kenya’s coral reefs: Decadal trends reveal

management needs

∗ 1

Melita A. Samoilys , Kennedy Osuka, George W. Maina , David O. Obura

CORDIO East Africa, P.O.BOX 10135, Mombasa 80101, Kenya

a r t i c l e i n f o a b s t r a c t

Article history: Kenya’s small scale coral reef ﬁsheries are extensively studied yet a practical understanding of the

Received 14 September 2015

resilience and status of the main target species remains largely elusive to the manager. We combined

Received in revised form 13 July 2016

a range of fishery and fish population descriptors to analyse Kenya’s coral reef fish and fisheries over

Accepted 27 July 2016

a 20 year period from the 1980s, to determine the sustainability of current ﬁshing levels and provide

recommendations for management. Fishers reported over 13 different artisanal ﬁshing gears of which

Keywords:

there are data for only the ﬁve widely used gears. Average catch rates declined 4-fold from the mid 1980s

Small-scale ﬁsheries

(13.7 ± 1.6 kg/ﬁsher/trip) to the 1990s (3.2 ± 0.1 kg/ﬁsher/trip) and then stabilized. Species richness in

Coral reefs

catches of these historically multi-species ﬁsheries declined dramatically and by 2007 only 2–3 species

Fisheries management

appeared in the top bracket (65–75% by number) with Siganus sutor (African whitespotted rabbitﬁsh) and

Leptoscarus vaigiensis (marbled parrotﬁsh) consistently being in this bracket in beach seine, gill net and

basket trap catches, contributing up to a maximum of 45% and 47% of the catch, respectively. Lethrinus bor-

bonicus dominated handline catches (50%). Relatively stable catch rates are reported from the 1990s to the

mid 2000s, likely maintained by shifting proportions of species in the catches. Patterns in ﬁsh population

densities over time show National Parks have helped increase densities of Lethrinidae and Haemulidae

and reduced the decline in densities of Scaridae and Acanthuridae, but that National Reserves have had

no positive effect. We suggest that the National Parks, which are No Take Zones (NTZs), and the ﬁsheries

regulations inside and outside of Reserves are inadequate for maintaining or restoring reef ﬁshery target

families under current levels of ﬁshery exploitation. We propose that recruitment overﬁshing of several

species and insufﬁcient areas under full protection, all exacerbated by climate change, are contributing to

driving Kenya’s artisanal coral reef ﬁsheries to a tipping point. We recommend species–speciﬁc manage-

ment options, changes in and enforcement of gear regulations and many more effective NTZs are needed

urgently if these ﬁsheries are to continue to provide livelihoods and food security on the Kenyan coast.

1. Introduction tence, both of which are poorly quantiﬁed (Obura 2001; Ochiewo,

2004; Cinner et al., 2009a). Consequently their contribution to

Small-scale artisanal coastal ﬁsheries can provide up to 99% national income and livelihoods is poorly acknowledged.

of the protein source to coastal households, provide over 80% of Small-scale artisanal coastal ﬁsheries are characterised as being

households’ income and therefore play a key role in food security multi-gear, multi-species and landed at multiple landings sites,

in developing countries (Barnes-Mauthe et al., 2013; Foale et al., as seen in Kenya (Kaunda-Arara et al., 2003; McClanahan and

2013; McClanahan et al., 2013). Despite this, they are frequently Mangi, 2004) and typical of many tropical ﬁsheries around the

undervalued by developing countries’ national policies (Henson world (Munro and Williams, 1985; Wright and Richards, 1985;

and Winnie, 2004; Hardman et al., 2013; Aloo et al., 2014). Further, Dalzell, 1996). This makes them difﬁcult to monitor and manage. If

artisanal ﬁshers operate largely to earn cash but also for subsis- monitoring is done it frequently only records total landings with-

out ﬁshing effort, making the data almost meaningless (Luckhurst

and Trott, 2009). Further, problems of overﬁshing and the use of

∗ destructive ﬁshing methods in these ﬁsheries are now widespread

Corresponding author.

and generally linked to poverty, over-population and poor gover-

E-mail addresses: [email protected]

(M.A. Samoilys), [email protected] (K. Osuka), [email protected] (G.W. Maina), nance (Allison et al., 2009; Gutiérrez et al., 2011). [email protected] (D.O. Obura).

Present address: The Nature Conservancy, Africa Regional Ofﬁce, Nairobi, Kenya

http://dx.doi.org/10.1016/j.ﬁshres.2016.07.025

178 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Coral reef fisheries are extensively studied small scale arti- to the first four of these locations. Sufficient data for long term CPUE

sanal ﬁsheries with many cases of over-exploitation (Russ 1991, trend analysis were only available for two locations: Mombasa and

2002; Newton et al., 2007), yet management reference points and Diani-Chale.

an applied understanding of the resilience of coral reef ﬁshes to A spatial overlay of nationally gazetted protected areas exists

exploitation remain largely elusive to the ﬁsheries manager on the (Fig. 1) comprising four Parks and ﬁve Reserves (Table 2). To exam-

ground. These ﬁsheries contribute a substantial portion of Kenya’s ine impacts of protective management we allocated data to one

artisanal catches (Government of Kenya, 2008, 2012) and also of three management zones: Parks (NTZs), Reserves and all other

represent one of the most studied in a developing country (Kaunda- areas called “Fished” which are not protected and are fully open

Arara et al., 2003; Mangi and Roberts, 2007; McClanahan et al., to ﬁshing. It should be noted that the enforcement of Parks and

2010; Hicks and McClanahan 2012) providing a valuable source of Reserves is variable, though Parks are considered well enforced

data and information. Yet our understanding of the status of these since the mid 1990s (Table 2; McClanahan et al., 2007).

ﬁsheries and deﬁning their most suitable management options

remains challenging. 2.2. Data collation

For decades researchers have tried to understand the con-

tribution of fin-fishes in terms of annual marine production or Published papers and some unpublished reports on fishing, fish-

yield from coral reefs (Marten and Polovina, 1982; Munro and eries and ﬁshes on the Kenya coast were reviewed to extract key

Williams, 1985; Russ, 1991; Dalzell, 1996). It is evident that reef variables on ﬁn-ﬁsh for assessing long term changes in Kenya’s arti-

−2 −1

yields >5 mt km yr are possible, with the highest productivity sanal ﬁsheries (Table 1). Catch per unit effort (CPUE) was used as

−2 −1

(>20 mt km yr ) reported from reefs in Philippines and Ameri- a measure of the state of each ﬁshery, where ﬁshery refers to an

can Samoa (Craig et al., 1993; Maypa et al., 2002). artisanal gear. We deﬁned as artisanal those gears used by local

In this study we examined Kenya’s coral reef ﬁsheries over two ﬁshers within territorial waters, limited to within 12 nautical mile

decades to provide practical management advice, to estimate their of the shore (GoK, 1989). These gears span those made from natural

annual yield and to contribute empirical evidence to recent debates ﬁbres that have been used traditionally for decades, to more mod-

on management options such as conventional gear and size limits ern gears involving man-made materials (Glaesel 1997; Samoilys

controls (Jennings et al., 2001; Hicks and McClanahan, 2012) and et al., 2011a).

spatial closures such as MPAs (Roberts and Polunin, 1991; Jennings, Papers that presented ﬁshery-independent measures of ﬁsh

2009). We sought to determine the sustainability of artisanal ﬁsh- population density from underwater visual census (UVC) surveys,

ing gears, their deployment and effort and also provide deﬁnitions were used to assess long term trends in population abundance of

of the gears. Speciﬁcally, we tested the following hypotheses that the species taken by artisanal gears, to provide a ﬁshery indepen-

are frequently stated by ﬁshery stakeholders in tropical ﬁsheries: i) dent measure of stock status.

artisanal ﬁshery catch rates are in decline and stocks are overﬁshed; A total of 23 papers and reports published between 1985 and

ii) populations of reef ﬁsh are declining. 2012, documenting ﬁsheries from 1984 to 2007 were used to

We used a series of meta analyses to combine a range of ﬁshery extract key CPUE and catch composition variables and indepen-

and ﬁsh population descriptors and parameters, both dependent dent UVC estimates of ﬁsh density (Table 1). Where values were

and independent of the ﬁshery, to assess the status of Kenya’s coral not speciﬁed in the paper but only presented graphically, values

reef ﬁsh and their ﬁsheries from the 1980s to the 2000s, based on were obtained using Data Thief lll software (Tummers, 2006) to one

published (e.g. Samoilys, 1988; Watson et al., 1996; Kaunda-Arara decimal point. A nine year (1998–2006) dataset (CORDIO unpubl.)

et al., 2003; McClanahan and Mangi, 2004; Mangi and Roberts, on artisanal fisheries (only data for fin-fish) in one location, Diani-

2007; McClanahan et al., 2007; McClanahan and Hicks 2011) and Chale, including UVC estimates of population abundance, was also

unpublished data (e.g. Carrara and Coppola, 1985; Coastal Oceans added and used to deﬁne certain parameters (see below). In most

Research and Development Indian Ocean (CORDIO) unpubl.). Multi- cases variables were estimated from means reported in papers and

ple data sources including estimates of ﬁsh population abundance therefore their precision may have been inﬂated.

that are independent of, but in combination with, ﬁsheries data

are important when trying to understand these complex ﬁsheries 2.3. Standardisation of variables and statistical analyses

(Connell et al., 1998; Daw et al., 2011). We examined trends in

catch rates and ﬁsh population abundance, species composition of Inevitably, reviewing a wide range of papers spanning many

catches and yields and juvenile retention rates of different gears to years encountered different methods, units and presentation of

understand gear impacts and the sustainability of current ﬁshing results. Therefore standardisation of variables was necessary. Some

levels. papers lumped all species together for UVC estimates of ﬁsh densi-

ties or CPUE, or aggregated locations and were therefore excluded

from the analyses. The following sections describe the parameters

2. Methods that were standardised and their statistical analyses. We cleaned

the data, ignored uncertain estimates or aggregated data and were

2.1. Study sites conservative in our calculations to minimise errors.

Kenya’s coral reefs occupy the shallow inshore zone, extending 2.3.1. Fishing gears and years

∼

offshore to <45 m depth, and at a distance of 0.5 km to 2 km from Based on sufﬁcient sample sizes across locations and years, data

the shoreline, except where river systems enter the sea (Obura et al., on ﬁve gears were analysed: large basket traps (∼6 cm mesh size),

2000). The coast is contained administratively within ﬁve Counties, gill net, handline, speargun and beach seine. Where possible data

formerly six Districts (Fig. 1). A wide range of study areas have been were extracted by year collected. Where data were not presented

reported, from small localised studies to the whole coast (Table 1). annually the median for the study period was taken to represent

Based on the geography of the coast the commonly reported study the year of the data.

sites were grouped into the following six locations (N to S): a)

Kiunga-Lamu, b) Malindi-Watamu, c) Mombasa, d) Diani–Chale, e) 2.3.2. CPUE

Gazi and f) Shimoni (Fig. 1). The four districts (Lamu, Malindi, Mom- The catch rate unit was standardised to kg/ﬁsher/day for each

basa, Kwale) reported by Carrara and Coppola (1985) were assigned of the five gears (one day is equivalent to one fishing trip – fishers

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 179

Fig. 1. Kenyan coast showing key locations, coastal administrative Counties and government gazetted marine protected areas. Insert shows the reef area estimated for

Diani-Chale for yield estimation.

180 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Table 1

Variables on Kenya’s artisanal coastal ﬁsheries extracted from published papers, reports and unpublished CORDIO data with corresponding study areas, years and data source.

Study Area refers to the study’s sites; Location in parenthesis refers to the locations in the present study to which the cited reference can be assigned. If the two correspond

only one name is given.

Parameter/question Variables Study Area (Location) Study Period Reference

CPUE (seasonal Gear CPUE Malindi to Watamu June 2000–February 2001 1. Kaunda and Rose (2004)

variation) (no.ﬁsh/ﬁsher/day) (Malindi-Watamu)

Malindi (Malindi-Watamu) January–December 2007 2. Locham et al. (2010)

CPUE (location Gear CPUE Diani-Chale 1998–2006 3. CORDIO unpublished data

variation over time) (kg/ﬁsher/day) Kenyatta, Mtwapa & Marina January 1998– October 1998 4. McClanahan and Mangi (2000)

(Mombasa)

Diani (Diani-Chale) 2003–2004 5. Mangi (2006)

Kenyatta, North coast (Mombasa) 1996–2005 6. McClanahan et al. (2008)

South coast (Diani-Chale)

Kwale, Mombasa, Malindi and 1984 7. Carrara and Coppola (1985)

Lamu Districts (Diani-Chale,

Mombasa, Malindi-Watamu,

Kiunga-Lamu)

Site CPUE Kenyatta (Mombasa), Diani-Chale September 1995–June 1999 8. McClanahan and Mangi (2001)

(kg/ﬁsher/day) Diani (Diani-Chale) 1995–2004 9. McClanahan et al. (2005)

Catch composition Gear catch composition Mombasa to Diani-Chale April 1998–August 2000 10. McClanahan and Mangi (2004)

by species/ﬁnﬁsh, Malindi-Watamu June 2000–February 2001 1 Kaunda-Arara and Rose (2004)

invertebrates, sharks (Malindi-Watamu)

and rays Malindi (Malindi-Watamu) January–December 2007 2 Locham et al. (2010)

Diani and Gazi (Diani-Chale) 1998–2006 3. CORDIO unpublished data

Mombasa June 2007–August 2007 11. McGuiness (2007)

Diani (Diani-Chale), Malindi and 1998–2006 12. McClanahan et al. (2010)

Watamu (Malindi-Watamu) &

Kisite (Shimoni)

Gear juvenile retention Mtwapa to Chale (Mombasa to May 2003- March 2004 13. Mangi and Roberts (2006)

Diani-Chale)

North coast (Mombasa) June 2007-August 2007 11. McGuiness (2007)

Fish densities Family and species Diani (Diani-Chale) 2001–2002 3. CORDIO unpublished data

density Parks, Reserves and Fished sites November 1987-March 1988 14. Samoilys (1988)

(Malindi-Watamu, Mombasa,

Diani-Chale, Shimoni)

Parks and Reserves (Shimoni) 1996 15. McClanahan et al. (1999)

Parks and Reserves (Shimoni) September–October 1992 16. Watson and Ormond (1994)

Parks and Reserves (Shimoni) January 1994–March 1994 17. Watson et al. (1996)

Parks and Fished sites from 1995–2001 18. McClanahan et al. (2002)

Vipingo-Diani (Malindi-Watamu,

Mombasa, Diani-Chale, Shimoni)

Parks and Fished sites 1992 19. McClanahan (1994)

(Malindi-Watamu, Mombasa,

Diani-Chale, Shimoni)

Parks and Reserves (Mombasa, 2004–2007 20. Munga et al. (2012) (KWS)

Malindi-Watamu, Shimoni)

Yield CPUE Diani-Chale 1999–2006 3. CORDIO unpublished data

Fish families yield Whole Coast (Malindi-Watamu, 1978–2001 21. Kaunda-Arara et al. (2003)

Mombasa, Diani-Chale, Shimoni)

Census of ﬁshers Diani-Chale 2003–2006 22. Tuda et al. (2008)

Fishing days per year Diani-Chale 2009–2010 23. Maina et al. (2013)

experimental ﬁshing.

Table 2

Marine protected areas in Kenya (Source: IUCN 2004) administered by the Kenya Wildlife Service (under the Wildlife Conservation and Management Act 1976; 2013). Date

established = gazettment. Park = fully protected, no ﬁshing allowed; Reserve = traditional ﬁshing gears permitted (all artisanal gears except illegal gears). Index of enforcement

is qualitative (MS, DO. pers. obs.).

Site IUCN Category Size (km ) Date established Protection status Relative index of enforcement

Malindi II 6.3 1968 Park Good

Watamu II 10 1968 Park Good

Malindi-Watamu VI 245 1968 Reserve Poor

Kisite II 28 1978 Park Good

Mpunguti VI 11 1978 Reserve Poor

Kiunga VI 250 1979 Reserve Poor

Mombasa II 200 1986 Park Good

Mombasa VI 10 1986 Reserve Poor

Diani-Chale VI 75 1995 Reserve None

very rarely do >1 trip in a day Obura, 2001; Maina et al., 2008). It experimental ﬁshing estimates, though they do, however, provide

should be noted that zero catches are rarely recorded and therefore good relative estimates. Catch rates were for all species aggregated;

all estimates of catch rate are likely to be higher than, for example, they were rarely reported by species or family.

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 181

Seasonal data were only available for basket traps as no. iii) Post-bleaching – (1999–2002) <5 years after the El Nino˜ event

individuals per species per trap per day (Table 1, papers 1,2). of 1997/8;

These data were used to assess seasonal differences in catch iv) Post bleaching – (2004–2007) >5 years after the El Nino˜ event

rates using the Mann Whitney U test (Zar, 1996) because data of 1997/8.

were non-normal after transformation. There was no signiﬁcant

difference (U = 4789.0, p = 0.896, n = 195) between the northeast Densities were compared between the four periods and three

monsoon (Nov-Mar; x = 0.172 0.051 SE), and the southeast mon- management zones (Park, Reserve, Fished). Our dataset did not

soon (Apr–Oct, x = 0.208 0.096 SE), We therefore pooled data extend beyond 2007 because other recent UVC studies could not

across the seasons for all subsequent analyses. be disaggregated by family or site, or surveyed different taxa from

The non-parametric Kruskal-Wallis test was used to compare those of our study. Data by the four locations (Malindi-Watamu,

catch rates between years and gears and the Mann-Whitney for Mombasa, Diani-Chale and Shimoni) were patchy, therefore loca-

post-hoc comparisons. It was not possible to compare Location tion differences were tested ﬁrst combining years and management

differences in CPUE, due to inadequate data for locations. zones and aggregating the 7 families, using the non-parametric

Kruskal-Wallis test (Zar, 1996). Locations were not signiﬁcantly

2.3.2.1. Converting location based CPUE to gear based CPUE. Two different (H = 4.064, p = 0.2546) and were therefore pooled for sub-

papers (McClanahan and Mangi, 2001; McClanahan et al., 2005) sequent analyses. Density of each family was compared across a

reported a lumped CPUE by landing site in Diani-Chale and Mom- combined factor of year periods and management zones (9 fac-

basa. This was converted to CPUE per gear based on gear catch tors) using the Kruskal-Wallis test followed by Mann-Whitney

rate proportions determined from Maina et al. (2008) for Diani and post-hoc test (Zar, 1996), and results graphed using box plots. Multi-

McClanahan and Mangi (2000) for Mombasa, as follows: dimensional scaling (MDS) using the Bray-Curtis similarity index

(Clarke and Warwick, 2001) was performed on three families with

Gear CPUE = Location CPUE × proportion of gear CPUE at location

sufﬁcient data (Scaridae, Acanthuridae, Siganidae) to test for differ-

ences in ﬁsh densities over time and between management zones.

The conversion was adjusted at three landing sites in Diani-Chale

A combined factor of management zone and year period was used

where beach seines were either never used, or were excluded from

and due to missing data, only 10 out of the 12 levels were possible.

1997 or 1998 onwards (based on McClanahan and Mangi, 2001).

In each MDS analysis a permutation-based analysis of similarities

(ANOSIM) was applied to identify differences in management zones

2.3.2.2. Converting kg/boat/day to kg/ﬁsher/day. To convert

and year periods (Clarke and Warwick, 2001). The ANOSIM on the

kg/boat/day CPUE per gear from Carrara and Coppola (1985)

Park sites showed that the two “Post-Bleach” year periods (<5yr

into kg/ﬁsher/day we divided the boat catch by mean number of

and >5 yr after 1997/8) were not statistically different (R = −0.081,

ﬁsher per gear (crew size) sourced from all data sets.

p = 0.634) and therefore were combined into one “Post-Bleach”

period for subsequent analyses.

2.3.3. Gear catch composition

Differences in species composition of catches between gears

2.3.5. Yield

were examined graphically over time. Mean% composition of

The average yield of ﬁn-ﬁsh on Kenya’s coral reef associated

species by gear was calculated from ﬁve papers (reference no.

coastal habitats was calculated for 1999 and 2006 from CORDIO’s

10,11,12,13,14, Table 1) including experimental ﬁshing studies.

catch landings data collected in Diani-Chale (see Maina et al., 2008

Percentages were used because most papers did not report abso-

for details of sampling). Total yield was calculated by the following

lute numbers of species per gear. Since these gears catch a very

equation (Russ, 1991):

large number of species (up to 163, McClanahan and Mangi, 2004),

analysis set an upper limit of those species that contributed cumu- C F FD

Y =

latively 65–75% of the catch to remove the infrequent species. A A 1000

range was necessary because species abundance cannot fall pre- Where:

cisely at 75%. The mean% catch per species per gear thus combined −1 −2

Y = annual yield (mt yr km )

species across studies. All papers on catch composition reported −1 −1

C = catch (kg ﬁsher day )

percentages of numbers of ﬁsh not weights of ﬁsh.

F = ﬁshing effort (mean total number active ﬁshers)

Catch composition for basket traps considered large mesh traps −1

FD = ﬁshing days per year (days yr )

only as small mesh traps are a relatively recent introduction with 2

A = area of reef ﬁshing grounds (km )

few data available. For basket traps, beach seines and gill nets data

The total area of reef-associated habitat of the Diani-Chale area

were also plotted by year to examine changes in species composi-

(see insert in Fig. 1) and the whole Kenyan coast was calculated

tion over time.

using a GADM base map of the coastline, with depth contours

from a digitised Admirality Chart (2002) and a layer of mangroves

2.3.4. Trends in ﬁsh densities

obtained from ReefBase (http://www.reefbase.org). Missing data

Underwater Visual Census (UVC) estimates of ﬁsh density of

from the Admirality Chart was supplemented by Google Earth

seven families (Lethrinidae, Lutjanidae, Haemulidae, Serranidae,

images. We calculated the area from shore to the 20 m-isobath line

Acanthuridae, Scaridae, and Siganidae) were obtained from pub-

using GIS. This covered a mix of coastal habitats including creeks

lished results (Table 1) and unpublished data (CORDIO data from

and bays, coral reefs and seagrass beds and some sandy lagoons but

Diani-Chale) and standardised to number/1000m . Since data were

did not include the mangrove forests or the large sand-mud Ung-

not available for every year, we aggregated data into four discrete 2

wana Bay. The total area for Diani-Chale was estimated as 56.4 km

periods which also corresponded with the coral bleaching event 2 2

and for the Kenya coast as 1724 km , excluding 54.3 km under

and impacts (Graham et al., 2007):

National Park (Fig. 1).

Information from a census of ﬁshers conducted in Diani-Chale

i) 1980s – earliest data, prior to strong management interventions

from 2003 to 2006 (Tuda et al., 2008) was used to provide an esti-

by government;

mate of the total number of ﬁshers in Diani-Chale that were active

ii) Pre-bleaching – (1992–1997) prior to the El Nino˜ event of

daily in 2006 and the mean number of ﬁshing days per year. Back

1997/8 which killed many of the corals in Kenya;

calculations on 2006 data were done to estimate total number of

182 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Table 3

ﬁshers for 1999. This should be treated as an approximation since

Results of analyses of similarity (ANOSIM) evaluating variation in densities of three

the time period is short for assuming a linear long term increase.

ﬁsh families across management zones and year periods.

Fishing days per year (FD) of 219 days was taken from a recent study −

R statistic p

in the Diani-Chale location (Maina et al., 2013).

Year periods

3. Results Global effect 0.233 0.004

1980s vs. Pre-bleaching 0.484 0.001

1980s vs. Post-bleaching 0.237 0.001

3.1. Fishing gears used on the Kenyan coast

Pre-bleaching vs. Post-bleaching 0.156 0.059

Management zones

Thirteen different fishing gears are used for fin-fish by artisanal

Global effect 0.211 0.001

ﬁshers with the gill net distinguished into 8 sub-types by ﬁsh-

Park vs. Fished 0.204 0.004

ers (Supplementary Table S1 in the online version at DOI: http:// Park vs. Reserve 0.329 0.002

−

dx.doi.org/10.1016/j.ﬁshres.2016.07.025). These gill net sub-types, Reserve vs. Fished 0.004 0.442

including the illegal monoﬁlament gill nets, were not distinguished

in any datasets, including government frame surveys, and therefore

(Fig. 4) which differed signiﬁcantly (Kruskal-Wallis test: H = 105.08,

speciﬁc impacts of each type could not be determined. Five gears

p < 0.001). These were data pooled from 1994 to 2006 for three

(basket trap, gill net, handline, speargun and beach seine) domi-

locations: Diani-Chale, Mombasa and Gazi. Speargun catch rates

nated in terms of published data and usage. These ﬁve gears were

(X = 3.9 ± 0.1 SE) were signiﬁcantly higher (p < 0.01) than all other

used only by men. The State Department of Fisheries (SDF)’s frame

gears and beach seine catch rates (X = 2.4 ± 0.1) were signiﬁcantly

survey reported four gears with more than 1000 pieces: basket

lower (p < 0.01) than all other gears (Fig. 4), whereas catch rates of

traps (3169), gill net (3956), handline (4132) and speargun (1007)

gill nets, handlines and basket traps were not signiﬁcantly different

(GoK, 2008). The beach seine had the lowest number (139) which

from each other (p > 0.05; Mann-Whitney tests). Each gear followed

could be attributed to its illegal nature, though the speargun, also

a similar pattern of peaks and troughs over the years (Fig. 4).

illegal, was well represented.

3.2. Catch species composition by gear 3.4. Trends in ﬁsh densities

The species composition varied between gears but two species Yearly averages of ﬁsh densities for the seven families revealed

dominated catches: Siganus sutor (African whitespotted rabbitﬁsh) high variability in the UVC data and several gaps with the most com-

and Leptoscarus vaigiensis (seagrass parrotﬁsh) appearing in the top plete dataset coming from Shimoni (Supplementary Graphs: Fig.

six species by number for all gears except handlines, contributing S1 in the online version at DOI: http://dx.doi.org/10.1016/j.ﬁshres.

up to a maximum of 44.8% (S. sutor) and 47.3% (L. vaigiensis) of 2016.07.025). Analyses were therefore constrained (see Meth-

the catches (Fig. 2). Thus, although gears generally differ in design ods) but nevertheless detected signiﬁcant differences in median

and deployment their primary catches were remarkably similar. ﬁsh densities between year periods and management zones for

The exception was the handlines where catches were dominated four of the seven families: Lethrinidae, Haemulidae, Acanthuridae

by lethrinids, notably Lethrinus borbonicus (49.9%). The greatest and Scaridae (Fig. 5). Patterns varied between these four fam-

diversity among the dominant species in the catches was seen ilies: lethrinid densities were signiﬁcantly lower in the 1980s

in handline and basket trap catches, and the least in speargun compared to Park sites in the Pre-bleaching (1992–1997) and com-

catches (Fig. 2). The number of dominant species in the catch has bined Post-bleaching (1999–2007) year periods (Fig. 5: p < 0.05),

diminished over years in three gears: basket traps, beach seine suggesting lethrinid abundance has improved over time in Parks

and gill net. Further, the relative contribution of target species has but not Reserves. The haemulid dataset is limited to 1980s and

increased 3 fold from 1999 to 2007 in beach seines (L. vaigiensis) Post-bleaching periods but also showed a signiﬁcant increase in

and basket traps (S. sutor) (Fig. 3). density in Park sites in the Post-bleaching period (Fig. 5). The

herbivorous acanthurids and scarids showed different patterns:

3.3. Trends in CPUE acanthurid densities were highest in the 1980s, where densities

did not differ between Fished, Reserve and Park sites. Densities

Catch rates for the whole coast, all gears combined, declined dropped signiﬁcantly in the Pre-bleaching period and continued to

4-fold from the mid 1980s (average of 13.7 1.6SE kg/ﬁsher/trip) drop in the Post-bleaching period (Fig. 5). In both later periods den-

to the 1990s (3.2 0.1SE kg/ﬁsher/trip). Records from the early sities were signiﬁcantly higher in Park sites compared with Fished

years are however only represented by one study (Carrara and Sites. Changes in scarid densities over time were highly variable,

Coppola, 1985): there are no data from 1985 to 1993. Catch with the highest densities seen in the Reserve sites in the 1980s.

rates from 1994 to 2006, with a mean of 3.2 ± 0.1SE kg/ﬁsher/trip, Densities did not differ between the 1980s and subsequent year

did not range widely (Fig. 4) but there were signiﬁcant differ- periods except between Fished sites in the 1980s and Park sites in

ences between years (Kruskal Wallis test: H = 77.22, p < 0.001). the Pre-bleaching and Post - bleaching periods where densities had

Pairwise comparisons showed differences which could be increased. These latter periods also differed signiﬁcantly in Parks

grouped into: i) two peaks, 2006 (4.9 ± 0.6SE kg/ﬁsher/trip) and with densities lower during the Post-bleaching period (Fig. 5).

1995–1997 (3.5 ± 0.1SE kg/ﬁsher/trip); ii) a trough, 1998–2000 Comparing management zone and year period simultaneously

(2.7 ± 0.1SE kg/ﬁsher/trip); iii) and moderate level catch rates in through the MDS plot provides a more powerful analysis of

2001–2005 (3.3 ± 0.1SE kg/ﬁsher/trip) excluding 2003 (3.0 ±0.1SE changes in ﬁsh densities over time and in relation to manage-

kg/ﬁsher/trip). This dataset was only available for two locations: ment zones, though sufﬁcient data were only available for the

Mombasa (average 2.9 ± 0.1SE kg/ﬁsher/trip) and Diani-Chale herbivorous scarids, siganids and acanthurids. There is a clear

(3.2 ± 0.1SE kg/ﬁsher/trip) (Fig. 4). Data from Gazi just south of clustering of sites surveyed in the 1980s, particularly for the

Diani-Chale showed higher catch rates during the 2000s (aver- Fished sites (Fig. 6). The ANOSIM revealed signiﬁcant differences

age 4.3 ± 1.6SE (2005) to 5.3 ± 1.5SE (2004) kg/ﬁsher/trip across all between year periods (Table 3) with 1980s abundance signiﬁ-

5 gears). Differences between years and locations were affected cantly different from Pre-bleach and Post-bleach periods. Park sites

by the relative proportions and catch rates of the ﬁve gears in the Pre-bleaching and Post-bleaching periods also clustered

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 183 Basket trap Handline

70 70 s sutor an u tj 60 60 Lethrinus borbonicus us a us Sigan 50 at i 50 n is i n sen i s mis n

40 40 er e bundance (%) ethr s sm n s i L mah s u Lethrinus len 30 u 30 in icu vaigi r hrinus n t Leptoscarus h o t Lethrinus mahsena Lethrinus lentjan

thrin 20 e b

20 Lutjanu Le e L fulviflamma xanthochilus Cheilios L 10 bor 10 Mean relative a

0 0 Beach seine 70 Gillnet 70

60 s s 60 u r si a 50 n n e ak 50 r ified a tosca Siganus sutor igi c a hse 40 k v tjan

a 40 a us pe Lep r n s s m sis u

30 n ed s

g 30 equula Un la sp. shar hurus ifi e t gie u i Leiognathus rinu

20 n i Lethrinus h 20 inel Siganus sutor va th Lethrinus mahsena triost hr Leptosca e Acan L 10 et Unspec 10 Lethrinus le Sard L Mean relative abundance (%) 0 0 Species

Speargun 70

60 vaigiensis Leptoscarus 50 d 40 ecifie s us 30 p m inu l iganus sutor Uns S 20 ato ro l a a n relative abundance % c a 10 C Me 0

Species

Fig. 2. Mean catch composition by gear with years and locations pooled, showing the dominant species − those that contribute up to 65% but not exceeding 75% of the catch

by number. Note some studies used synonyms identifying L. lentjan as L. mahsenoides, and L. mahsena as L. sanguineus, these are updated to the senior synonym here. Error

bars are SE. (Data Source: 5 studies, Table 1).

−2

together, while Reserve sites in Pre- and Post-bleaching periods intensity of 4.0 ﬁshers km each day. Using the Yield equation (see

were widely scattered in different quadrants (Fig. 6). Management Methods), these estimates gave a mean yield for Diani – Chale of

−1 −2 −1 −2

zones differed signiﬁcantly and pairwise comparisons revealed 2.29 mt yr km in 1999 and 6.09 mt yr km in 2006. This rep-

Parks differed signiﬁcantly from Fished and Reserve (Table 3) indi- resents an annual increment of 354 kg per year (y = 0.354x + 2.555,

cating ﬁsh densities in Reserves are similar to Fished sites. The R = 0.70).

bi-plot overlay of the direction of change in the three families’ abun-

dance shows highest acanthurid densities associated with sites

4. Discussion

surveyed in the 1980s and the reverse in the scarids with highest

densities in both later year periods in the Parks; these changes are

Using a series of meta-analyses as well as ﬁeld observations on

reﬂected in signiﬁcant Mann-Whitney tests (Fig. 5). The siganids

ﬁshing gear utilization to assess Kenya’s artisanal coral reef ﬁsh-

mirrored the acanthurids but the change was weak and not signif-

eries we found 13 different artisanal gears. However, gill nets are

icant (Fig. 6).

only reported as one gear yet their sub-types span >40 cm in mesh

size. Their different catches therefore remain unquantiﬁed which

3.5. Coral reef ﬁsheries yield is a concern since ﬁshing offshore with gill nets for more abun-

dant pelagic species is often recommended to address dwindling

Catch per unit effort (CPUE) for Diani-Chale increased signif- demersal stocks close to shore (Bell et al., 2009).

−1 −1 −1 −1

icantly from 2.9 kg ﬁsher day in 1999 to 5.4 kg ﬁsher day

in 2006 (y = 0.228x + 3.100, R = 0.56). Fishing effort (total number 4.1. Trends in catch per unit of effort (CPUE)

of active ﬁshers per day) in Diani-Chale ranged from 264 in 2003

to 292 in 2006, representing a mean annual increment of 3.6%, CPUE is often assumed to be proportional to ﬁsh abundance and

∼

equivalent to an increase of 7 ﬁshers per year, though this should is one of the most commonly used indices of stock abundance in

be treated as an approximation since the increase was not uni- ﬁsheries science (Richards and Schnute, 1986). Average catch rates

form each year and the time period is short for assuming a linear in Kenya in 2006 were 23% of what they were in the mid 1980s.

long term increase. With this assumption, these values allow back- Artisanal coral reef based ﬁsheries typically yield around 25% of

calculated figures for 1999 of 228 daily active fishers and fishing their original or “virgin” catch rates (Jennings and Polunin, 1995,

184 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 i) Basket trap 80 80 2000 1999

70 70 utor s a

60 60 a n se or ganus i t an 50 h 50 j u t S n

s 40

40 u sma u n

30 30 ri h Siganus s et Lethrinus mahsen L

20 vaigiensis 20 Lethrinus le Leptoscar Lethrinus miniatus 10 10 Mean relative abundance (%)

0 0 s

80 2002 80 u 2007 n a g 70 70 sutor Si 60 60 iganus sutor

ndance (%) 50 50 S s a n

40 nsis 40 us s hse hrinu us t n u a i oides in 30 r 30 n Le m vaigie nic Leptoscarus o ahsena ethr thrinus b 20 Leth 20 L m e ahse or L m b ean relative abu 10 10 M

0 0 ii) Gill net 70

70 19 99 2007 60 60 sutor s

50 u

50 n us sutor a g k ed Si

40 a r

40 ifi Sigan a c abundance (%) s pe u r ve 30 i 30 s ns t a nus h us i c r a si U arus s u n c el h e r to igiensis i 20 t 20 os p a g n Lethr i n v ostegus a i a Le v tr ea Ac 10 10 Lept M

0 0 iii) Beach seine

u 70 1 999 70 2007 sis ar n 60 e

%) 60 igi tosc a p e( ula v e c u 50 50 L an sena h nd r s

40 o 40 s n t i hus eq ru t u k a a s . a ntja p ens

s 30

30 ar ed le u tive abu i h iogn n a la s s l a el vaigi Le r 20 ethrinus ma

20 Leptosc Siganus sutor nu ne pecif L Sig ri s h rdi n t

a 10 e 10 U Mean Lethrinus S L

0 0

Species Species

Fig. 3. Decline in number of species in catches over time from basket traps, gill nets and beach seine (n = 2-5 studies, see Table 1).

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 185

Diani - Chale Basket trap

Beach seine

Gillnet

K enyan coast 7 Handline Speargun 6

9 Basket traps 5 Beach seine 8 Gillnet 4 Handline Speargun 3 7

CPUE (kg/fisher/day) 2

6 1 1997 1998 1999 2000 1995 2005 2006 1994 1996 2001 2002 2003 2004 5 Years Mombasa

Basket trap

Beach seine

6 Gi llnet 4 H andline Speargun 5 CPUE (Kg/fisher/day) CPUE (kg/fisher/day) 3 4

3 2 2

1 CPUE (Kg/fisher/day) 1 4 8 4 02 96 9 92 9 0 00 9 9 9 0 9 2 2006 2008 1 2000 2 1 1 1 1993 1994 1995 1999 2001 2002 1996 1998 2004 2005 1997 2000 2003 2006

Years Years

Fig 4. Average CPUE values (kg/ﬁsher/day) by year for the ﬁve most common gears, locations pooled. Inserts show two locations separated. Error bars are standard errors.

(Data Source: Kenyan coast − 6 studies, Diani − 5 studies, Mombasa −3 studies, see Table 1).

1996). However, we know the 1980s CPUE cannot represent virgin ever, by 2007 only 2–3 species appeared in the top bracket for any

catch rates since artisanal ﬁshing has occurred on Kenya’s coast for gear (using the top 65–75% number of ﬁsh in the catch) with S. sutor

several hundred years (Glaesel, 1997). This suggests the mean value consistently being in this bracket in beachseine, gill net and basket

of ∼3 kg per ﬁshing trip from 1994 to 2006 across all gears is low. trap catches, and L. vaigiensis in gill net and beach seine catches and

Although the 1980s value of 14 kg/ﬁsher/day is based on only one Lethrinus borbonicus in handline catches. Similarly, on the southern

study (Carrara and Coppola, 1985), similar catch rates are known Kenyan coast only 15 species contributed 90% of the catch with S.

from coral reef ﬁsheries elsewhere in the 1980s and early 1990s. sutor, L. vaigiensis and Lethrinus lentjan being the three most com-

For example, upper CPUE values from a range of ﬁshing gears on monly caught species (Hicks and McClanahan, 2012), and studies

reef and lagoon ﬁsheries reviewed by Dalzell (1996) ranged from in Madagascar have similar ﬁndings (Davies et al., 2009). This dra-

19 kg to 32 kg per set in Fiji, to 43 kg per trip in Kiribati, while Craig matic shift in species composition illustrates that Kenya’s artisanal

et al. (1993) report up to 21 kg per day in American Samoa in 1991. ﬁsheries can no longer be characterised as highly multi-species and

Aggregated catch rates in the 1990s and 2000s showed two that some species have gone commercially extinct.

peaks: in 1995–1997 and 2006 (3.5 & 4.9 kg/ﬁsher/trip, respec- Our understanding of the resilience of S. sutor and L. vaigiensis

tively); and a minimum value of 2.7 kg/ﬁsher/trip in 1998–2000. to this increase in ﬁshing pressure remains poor. Recent mortality

These ﬂuctuations might relate to the 1997/8 coral bleaching event, estimates for these two species in lagoonal ﬁsheries on the south

with recovery suggested by catch rates averaging 3.3 kg/ﬁsher/trip coast reveal both are overexploited (Hicks and McClanahan, 2012).

during 2001–2005. This interpretation is supported by the con- Other factors related to their biology and their market niche may

sistent differences in gear CPUE which matched these same also render S. sutor vulnerable to overﬁshing. They form spawning

ﬂuctuations. Our results found speargun catch rates were signif- aggregations which are targeted by ﬁshers making them poten-

icantly higher and beach seine catch rates signiﬁcantly lower than tially vulnerable to local extirpation, though this species with short

all other gears. This contrasts with an artisanal ﬁshery in Madagas- life span and rapid growth is particularly resilient to ﬁshing pres-

car where beach seine catch rates are reported to be the highest sure (Robinson et al., 2011; Samoilys et al., 2013). In addition, S.

(Davies et al., 2009). Of concern is that the catch rates over the sutor is a key target species in women’s (mama karanga) small

12 year period from 1994 may have been artiﬁcially maintained by scale ﬁsh trade (Karuga and Abila, 2007; Samoilys and Tuda, 2009)

shifting proportions of species in the catches, with S. sutor and L. which has been suggested as a driver for the persistent use of

vaigiensis increasingly contributing to the catch. illegal beach seines (McClanahan and Mangi, 2001; Harrison and

Samoilys, 2009; McClanahan et al., 2008), though no such link

was detected between ﬁsher targets and women’s small scale ﬁsh

4.2. Species level changes and management implications

trade in Philippines (Pomeroy et al., 2016). Concerns over the sus-

tainability of current ﬁshing pressure on S. sutor and L. vaigiensis

Reef ﬁsheries are well known for being highly multi-species in

led to a Policy Brief to government in 2011 recommending full

character (Polunin and Roberts, 1996): typical catches in Kenya

stock assessments and focussed management plans for these two

include over 160 species (McClanahan and Mangi, 2004). This char-

species (Samoilys et al., 2011b), and full enforcement of the illegal

acteristic can lead to inferences of resilience at the ﬁsh community

beach seine. A similar recommendation to reduce beach seine use

level since ﬁshing pressure is spread across a wide range of species.

comes from Madagascar’s artisanal ﬁsheries (Davies et al., 2009).

Here we deﬁne resilience as the ability of the reef ﬁsh community

to resist, or absorb and recover its structure and functions. How-

186 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Fig. 5. Median densities (with 25%, 75% quartiles) pooled for the whole coast of four ﬁsh families that differed signiﬁcantly across year periods and management zones.

Medians are calculated from 4 to 7 studies per family (see Table 1). Plots with the identical lowercase letters are not signiﬁcantly different based on Mann-Whitney post-hoc

test. * = data not collected. Open dots indicate outliers not included in median calculations.

Our results belie the notion that the multi-species nature of Kenya’s Combining the results of gear speciﬁc CPUE rates, changes in

coral reef ﬁsheries makes them more resilient to ﬁshing pressure. species composition of catches and juvenile retention rates (Mangi

Lethrinids are a valuable food ﬁsh group and range across a wide and Roberts, 2006; McGuiness 2007) leads us to recommend

variety of habitats, not just coral reefs (Carpenter and Allen, 1989; the following management options for these ﬁsheries: a) prop-

Connell et al., 1998; Fitzpatrick et al., 2012). It is therefore of con- erly enforce the ban on beach seines particularly in the National

cern that they were absent in the 65–75% bracket of most recent Reserves; b) trial a hook size limit to reduce juvenile retention of

catches (2007) of beach seines and gill nets. Analysis of UVC data lethrinids in handline ﬁshing; c) lift the ban on spearguns because

showed that lethrinid abundance has increased in Parks in recent they showed the lowest retention rate of juveniles, have no by-

years suggesting NTZs on coral reefs can provide a positive impact catch and have low input costs for young ﬁshers starting their

on these taxa. Since beach seine catches comprise 68% juveniles ﬁshing livelihood; d) review and test minimum and maximum

(Mangi and Roberts, 2006), proper enforcement of the ban on this mesh size in gill nets to protect juveniles and reduce capture of

gear will protect immature ﬁsh, but this requires ﬁshers’ coopera- Endangered or Vulnerable mammals, sharks and turtles; and e) con-

tion in policing. In addition, gear regulations such as hook size and duct biological research on Siganus sutor and Leptoscarus vaigiensis

mesh size could be trialed to effect a minimum size limit (Hicks to complete parameters required for a full stock assessment. These

and McClanahan, 2012) in Fished areas and Reserves to help restore recommendations should be combined within a Protected Areas

lethrinid populations. management framework, which ensures, for example, that the

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 187

Fig. 6. PCA bi-plot of mean abundance of three families (Acanthuridae, Scaridae and Siganidae) overlaid on MDS plot of the same data from three year periods and three

management zones. (Data Source: 8 studies, Table 1).

largest individuals are protected thereby maximising fecundity With declining trends in ﬁsh densities combined with declines

(Bohnsack, 1996). in number of species in the ﬁshery catches, the sustainability of

current ﬁshing practises on Kenya’s coral reefs appears question-

4.3. Trends in fish densities able. Parks are probably insufficient as a fisheries management tool

since Kenya protects only 3% of its shallow coastal reef associated

Long term trends in ﬁsh population abundance were difﬁcult waters, whereas international targets are set at 10% (Convention

to discern partly due to highly variable and patchy UVC data, on Biological Diversity Aichi Target 11) and as high as 25–30%

nevertheless, signiﬁcant differences in abundance of Lethrinidae, (Fernandes et al., 2005). However, a positive development is that

Haemulidae, Scaridae and Acanthuridae were detected between coastal communities in Kenya are increasingly establishing Locally

the 1980s and the Pre-bleaching (1992–1997) and Post-bleaching Managed Marine Areas (LMMAs), which include NTZs, in response

(1999–2007) periods. Acanthurids and scarids were more abun- to catch declines and to exert ownership of their resources (Rocliffe

dant in the 1980s compared with both later periods, whereas et al., 2014; Kawaka et al., 2015). We recommend this is encouraged

lethrinid and haemulid densities were lowest in the 1980s. Den- and that stronger management measures taken in the government

sities of all four families in Parks (equivalent to NTZs), were Reserves to improve their effectiveness.

generally higher than in Reserves and Fished sites, and the latter

two zones did not differ, suggesting Park management is having 4.4. Yield and value of Kenya’s reef ﬁsheries

a positive effect on ﬁsh densities whereas Reserve management

−2 −1

is not. Patterns in scarid densities suggest Parks may have mit- Yields (mt km yr ) from coral reef ﬁsheries vary enormously

igated a decline in scarid densities over years, though this was (Table 4) partly due to a range of variables that are difﬁcult to

reduced by a negative impact from the 1998 bleaching. It is likely measure accurately (Russ, 1991; Arias-Gonzalez et al., 1994) and

that Parks originally had higher coral cover and therefore would the problematic assumptions in comparing multi-species ﬁshery

be more impacted by bleaching. The patterns suggest that Parks yields (Koslow et al., 1994; Russ and Alcala, 1998; Jennings et al.,

are associated with either increasing ﬁsh population densities or 2001) all issues met in the current study. Variation in reference area

reducing overall declines in population densities. Increased pop- size is seen in previous estimates from Kenya which range from

2 2

ulation densities (lethrinids and haemulids) after the 1980s may 620 km to 1080 km (Sheppard and Wells, 1988; McClanahan and

relate to government improvements in protection and manage- Obura, 1995; Kaunda-Arara et al., 2003; Muthiga et al., 2008). Our

ment of reef habitats and new ﬁsheries regulations (Samoilys, 1988; larger area estimate (which gave a smaller ﬁshery yield) includes

Government of Kenya 1991, 2001; Watson et al., 1996; McClanahan additional shallow-water habitats such as sandy lagoon, mangrove

et al., 2007). Declines in densities of lethrinids and scarids in fringes and sea grass beds, as these are habitats in which arti-

Parks Post-bleaching is likely to reflect impacts from the 1998 sanal fishers range to catch reef-associated fin-fishes (Bellwood

coral bleaching event (Graham et al., 2011). However, we found and Wainright, 2002). Moreover, when species compositions of

no evidence for a lag effect of around six years (Graham et al., catches change over time, as seen here, ﬁshery productivity will

2007): ﬁsh densities did not differ between <5 years and >5 years also change. Nevertheless, yields provide a perspective on these

post bleaching. No clear explanation was found for the decline in complex ﬁsheries that can help to understand the state of the ﬁsh-

acanthurids over the 20 year period. The lack of increase in ﬁsh ery (Russ, 1991) and provide a guide for government. The annual

−1 −2

densities in Fished and Reserve sites after the 1980s and much yield of 6 mt yr km , estimated from Diani-Chale in 2006, is rel-

larger declines in densities in these zones compared to Parks by atively high compared with typical global yields but is within the

the Post-bleaching period, suggest that the ﬁsheries management ranges reported previously in Kenya, with two higher exceptions

−1 −2

regulations of Reserves and Fished sites are inadequate. (Table 4). The highest yields, >15–20 mt yr km , are seen from

188 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Table 4

−2 −1

Fisheries yield in metric tonnes per unit area per year (mt.km year ) of demersal fin-fish obtained in different coral reef areas in the Western Indian Ocean, the Pacific

Islands, Asia and Caribbean. Most recent yields presented ﬁrst. n.a. = year not speciﬁed, publication date provides an indication.

−2 −1

Location mt km year Year Habitat Source

Western Indian Ocean

Kenya

Diani-Chale 6.1 2006 reef This study

Diani-Chale 2.3 1999 reef This study

Mombasa and Diani-Chale 3.0–16.0 1996–2005 reef McClanahan et al. (2008)

Whole coast 2.1 1978–2001 reef Kaunda-Arara et al. (2003)

Mombasa and Diani-Chale 6.6 1995 McClanahan and Mangi

Mombasa and Diani-Chale 4.5 1999 reef and lagoon (2001)

Kiliﬁ (Malindi-Watamu) 8.8 1982–1984 reef Nzioka (1990)

Northern coast (Lamu & 4.9 1977 reef FAO (1979)

Malindi-Watamu)

Southern coast (Mombasa and 5.6 1977 reef FAO (1979)

Diani-Chale)

Tanzania− whole coast 4.8 1977 reef FAO (1979)

Seychelles − Mahe (East) 1.4 1977 reef FAO (1979)

Seychelles − Mahe (West) 3.1 reef FAO (1979)

−

Madagascar Tulear Region, 12.0 1991 barrier, lagoon and fringing Laroche and Ramananarivo

southwest reefs & coral banks (1995)

Mauritius 3.5 1977 coralline shelf FAO (1979)

Mauritius 4.7 reef Wheeler and Ommanney (1953)

Paciﬁc Islands

American Samoa − Outer 2.3 2002–2003 Craig et al. (2008)

Islands

American Samoa − Tutuila 21.2 1979 reef and reef ﬂat Wass (1982) Island

Fiji

Ono-i-Lau 2.9 2002 reefs Kuster et al. (2005)

Ono-i-Lau 3.7 1982 reefs Kuster et al. (2005)

Astrolabe reef 0.9 1992–1993 fringing reef Jennings (1998)

Yanuca, Dravuni, Moala, 10.2 1992–1993 fringing reef Jennings and Polunin (1995)

Totoya, Nauluvatu

Kiribati − Tarawa 4.4–7.2 n.a coral atoll (reefs and lagoon) Marriott (1984); Mees et al.

(1988)

Micronesia − Kapingamarangi 0.7 n.a coral atoll Stevenson and Marshall (1974)

Micronesia − Ifaluk Atoll 5.1 1962–1963 reef and lagoon Stevenson and Marshall (1974)

Micronesia − Lamotrek and 0.45/0.09 n.a reef Stevenson and Marshall (1974)

Raroi Atolls Asia

Philipines

Apo Island 15.0–20.0 1997–2001 reef Maypa et al. (2002)

San Salvador Island 7.0 14.0 1988/89 1989/90 reef Christie and White (1994)

Sumilon Island 10.1 1976–1986 reef Alcala (1981); Alcala and Russ (1990)

Caribbean

Bermuda 0.4 n.a reef Bardach and Menzel (1957)

Bahamas 2.5 n.a reef Gulland (1971)

Jamaica 2.0 n.a reef Munro (1969)

Papua New Guinea − Tigak 0.4 n.a Reef Wright and Richards (1985)

Islands

reef areas with >50% live coral cover and where herbivores and ues from Diani-Chale to the whole Kenyan coast gives a total yield

planktivores dominate catches such as in Philippines (Russ and of 7754 mt of reef-associated ﬁn-ﬁsh in 2006 valued at USD 9.69

Alcala, 1989; Maypa et al., 2002). Kenyan artisanal catches are now million (at KSh 100 per kg, USD 1 = KSh. 80), captured by 8682 ﬁsh-

dominated by the herbivorous siganids and scarids and average ers, slightly more than the annual production (6000–7000 mt) of

coral cover is around 20% (Ateweberhan et al., 2011), placing Kenya all marine ﬁsheries reported as an aggregate value by government

in the mid-range of productivity of reef ﬁsheries. We propose that (GoK, 2005, 2008). Our estimate is crude as it assumes that the pro-

the increase in yield at Diani-Chale from 1999 to 2006 does not ductivity of coastal habitats supporting these ﬁsheries are broadly

reflect a fishery below MSY (Jennings et al., 2001) but is likely to be similar along the coast and that fisher density is also similar, nei-

related to increasing modernisation of gear (Samoilys et al., 2011a), ther of which can be validated. In addition, it is an extrapolation

increasing use of nets, including the “ring net” and shifts in species from a small intensively ﬁshed reef area considered to be overex-

composition from carnivores and omnivores to herbivores. We sug- ploited (McClanahan and Obura, 1995; Obura, 2001). Nevertheless,

−1 −2

gest, therefore, that a yield of 6 mt yr km is a realistic maximum these calculations serve to illustrate the likely high value of Kenya’s

for Kenya’s reef fisheries, but caution that current management coastal artisanal fisheries and hence the investment benefits in

practices are unlikely to maintain this. managing these ﬁsheries for food security.

Small scale and artisanal ﬁsheries account for more than 40%

of the global marine ﬁsheries catch taken for human consumption

5. Conclusions

(Jennings et al., 2001). Kenya’s artisanal marine ﬁsheries are con-

tributing valuable ﬁsh landings and support the livelihoods of some

The long term impacts of ﬁshing on Kenya’s coral reef ﬁsh popu-

13,000 ﬁshers (GoK, 2012) on the coast. Extrapolating the yield val-

lations is easily hidden by aggregated catch data. Complacency over

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 189

the state of these fisheries may be triggered by increasing fishery Barnes-Mauthe, M., Oleson, K.L., Zafindrasilivonona, B., 2013. The total economic

value of small-scale ﬁsheries with a characterization of post-landing trends: an

yields within the range expected for relatively healthy coral reefs

application in Madagascar with global relevance. Fish. Res. 147, 175–185.

and the knowledge that Kenya’s Marine Parks are effective. We sug-

Bardach, J.E., Menzel, D.W., 1957. Field and laboratory observations on the growth

gest an alternative picture where shifts in species composition of of some Bermuda reef ﬁshes. Proc. Gulf Caribb. Fish Inst. 9, 106–112.

Bell, J.D., Kronen, M., Vunisea, A., Nash, W.J., Keeble, G., Demmke, A., Pontifex, S.,

catches are contributing to very high pressure on just a few species,

Andréfouët, S., 2009. Planning the use of ﬁsh for food security in the Paciﬁc.

causing recruitment overﬁshing (Pauly, 1988). The high juvenile

Mar. Policy 33, 64–76.

capture of S. sutor, Lethrinus mahsena and L. lentjan by all the gears of Bellwood, D.R., Wainright, P.C., 2002. The history and biogeography of ﬁshes on

coral reefs. In: Sale, P.F. (Ed.), Coral Reef Fishes: Dynamics and Diversity in a

the artisanal ﬁshery (>75%, Mangi and Roberts, 2006), the increas-

Complex Ecosystem. Academic Press, San Diego, California, USA, pp. 5–32.

ing numbers of ﬁshers entering the ﬁshery each year (GoK, 2012),

Bohnsack, J.A., 1996. Maintenance and recovery of reef ﬁshery productivity. In:

and inadequate protection from Parks (too small) and Reserves (not Polunin, N.V.C., Roberts, C.M. (Eds.), Reef Fisheries. Chapman & Hall, London,

well enforced) for some key ﬁshery taxa, are all putting populations pp. 283–313.

Carpenter, K.E., Allen, G.R., 1989. FAO Species Catalogue. Emperor Fishes and

at risk from over-exploitation. Since coral reef associated ﬁshes

Large-eye Breams of the World (family Lethrinidae). An Annotated and

are also highly susceptible to climate change (Graham et al., 2011;

Illustrated Catalogue of Lethrinid Species Known to Date. FAO Fisheries

Kroeker et al., 2013) we suggest the artisanal coral reef ﬁsheries of Synopsis. No. 125, vol. 9. FAO, Rome, 118pp.

Carrara, G., Coppola, R.S., 1985. Results of the First Year of Implementation of the

Kenya may be approaching a tipping point. To safeguard the future

Kenyan Catch Assessment Survey RAF/79/065/WP/18/85/E, 85 pp.

of Kenya’s coral reef ﬁsh populations, as well as the ﬁshers depen-

Christie, P., White, A., 1994. Reef ﬁsh yield and reef condition for San Salvador,

dent on them, we recommend a combination of species–speciﬁc Luzon, Philippines. Asian Fish. Sci. 7, 135–148.

Cinner, J.E., Daw, T.M., McClanahan, T.R., 2009a. Socioeconomic factors that affect

management options, changes in gear regulations, and more NTZs

artisanal ﬁshers’ readiness to exit a declining ﬁshery. Conserv. Biol. 23 (1),

through better enforced Reserves and community LMMAs. Such 124–130.

mechanisms should improve the resilience of ﬁsh populations and Clarke, K.R., Warwick, R.M., 2001. Changes in Marine Communities: an Approach to

Statistical Analysis and Interpretation. PRIMER-E Ltd., Plymouth, 176p.

coral reefs to the combined and interacting effects of ﬁshing and

Connell, S.D., Samoilys, M.A., Lincoln Smith, M.P., Leqata, J., 1998. Comparisons of

climate (Perry et al., 2010).

abundance of coral-reef ﬁsh: catch and effort surveys versus visual census.

Aust. J. Ecol. 23 (6), 579–586.

Craig, P., Ponwith, B., Aitaoto, F., Hamm, D., 1993. The commercial, subsistence, and

recreational ﬁsheries of American Samoa. Mar. Fish Rev. 55 (2), 109–116.

Author contributions

Craig, P., Green, A., Tuilagi, F., 2008. Subsistence harvest of coral reef resources in

the outer islands of American Samoa: Modern, historic and prehistoric catches.

MS – initial concept, methods, data analysis oversight and Fish. Res. 89, 230–240.

interpretation, results interpretation and lead author on writing Dalzell, P., 1996. Catch rates, selectivity and yields of reef ﬁshing. In: Polunin,

N.V.C., Roberts, C.M. (Eds.), Reef Fisheries. Chapman and Hall, London, pp.

manuscript; KO – sourcing data, data analysis and interpretion,

161–192.

inputs to writing; GM – some data sourcing and analysis and inputs

Davies, T.E., Beanjara, N., Tregenza, T., 2009. A socio-economic perspective on

to writing; DO – data collection and data interpretation. gear-based management in an artisanal ﬁshery in south-west Madagascar.

Fish. Manag. Ecol. 16 (4), 279–289.

Daw, T.M., Robinson, J., Graham, A.J., 2011. Perceptions of trends in Seychelles

artisanal trap ﬁsheries: comparing catch monitoring, underwater visual census

Acknowledgements

and ﬁshers’ knowledge. Environ. Conserv. 38 (1), 75–88.

FAO. 1979. Report on the FAO/IOP Workshop on the Fisheries Resources of the

Western Indian Ocean South of Equator. In: J. Gulland (Ed.) IOFC/DEV/79/45

This work was funded through the United States Agency

FAO, Rome.

for International Development (USAID) and implemented under

Fernandes, L., Day, J.O.N., Lewis, A., Slegers, S., Kerrigan, B., Breen, D.A.N., et al.,

Research Permit Number NCST/RRI/12/1/BS/205 under National 2005. Establishing representative no-take areas in the great barrier reef:

large-scale implementation of theory on marine protected areas. Conserv.

Council for Science and Technology, Kenya. USAID had no involve-

Biol., 1733–1744.

ment in the research. We thank the following ﬁshers and BMU

Fitzpatrick, B.M., Harvey, E.S., Heyward, A.J., Twiggs, E.J., Colquhoun, J., 2012.

members who gave us valuable information on ﬁshing gears: Habitat specialization in tropical continental shelf demersal ﬁsh assemblages.

PLoS One 7 (6), e39634, http://dx.doi.org/10.1371/journal.pone.0039634.

Buruku Hamis Hamsa, Hussein Ali Mwabori, Hemedi Mohammedi

Foale, S., Adhuri, D., Alino,˜ P., Allison, E.H., Andrew, N., Cohen, P., 2013. Food

Mwafujo, Ali Shekue, Hamisi Kirauni, Salim Sadik, Ahmed Sheb-

security and the Coral Triangle initiative. Mar. Policy 38, 174–183.

wana and Emmanuel Yaa, and the following for ﬁeld assistance: Glaesel, H., 1997. Rough Waters: Community and Ecological Change near Kenya’s

Mbaraka Hemedi, Ali Mwabori and Pablo Samoilys. We are also Marine Parks. PhD Dissertation (Geography). University Wisconsin, Madison.

Government of Kenya, 1989. Maritime Zones Act Cap. 371.

grateful to all those who collected data in the studies cited, notably

Government of Kenya, 2005. Annual Report. Ministry of Livestock and Fisheries

KWS. We are grateful to Dr Boaz Kaunda-Arara for his critical review

Development. Fisheries Department, Republic of Kenya.

of an early draft, and to the helpful suggestions from three anony- Government of Kenya, 2008. Frame Survey Report. Ministry of Fisheries

Development, Kenya. Provincial Headquarters, Mombasa, 143 pp.

mous reviewers.

Government of Kenya, 2012. Marine Waters Fisheries Frame Survey Report.

Ministry of Fisheries Development, Department of Fisheries, Nairobi, 85 pp.

Government of Kenya, 1991. Fisheries Act, Cap 378.

References Government of Kenya, 2001. Kenya Gazette Notice No. 7565 Vol. CIII. No. 69 of 9th

November 2001.

Graham, N.A., McClanahan, T.R., Graham, N.A.J., Wilson, S.K., Jennings, S., Polunin,

Alcala, A.C., 1981. Fish Yield of Coral Reefs of Sumilon Island, Central Philippines.

N.V.C., Robinson, J., Bijoux, J., Daw, T.M., 2007. Lag effects in the impacts of

NRCP Res. Bull. 36, 1–7.

mass coral bleaching coral reef ﬁsh, ﬁsheries, and ecosystems. Conserv. Biol. 21

Alcala, A.C., Russ, G.R., 1990. A direct test of the effects of protective management

(5), 1291–1300.

on abundance and yield of tropical marine resources. J. Cons. Int. Explor. Mer.

Graham, N.A.J., Chabanet, P., Evans, R.D., Jennings, S., Letourneur, Y., MacNeil, M.A.,

47 (1), 40–47.

et al., 2011. Extinction vulnerability of coral reef ﬁshes. Ecol. Lett. 14, 341–348.

Allison, E.H., Perry, A.L., Badjeck, M.C., Adger, W.N., Brown, K., Conway, D., Halls,

Gulland, J.A., 1971. The ﬁsh resources of the ocean. Fishing News Books, West

A.S., Pilling, G.M., Reynolds, J.D., Andrew, N.L., Dulvy, N.K., 2009. Vulnerability

Byﬂeet, UK.

of national economies to the impacts of climate change on ﬁsheries. Fish Fish.

Gutiérrez, N.L., Hilborn, R., Defeo, O., 2011. Leadership, social capital and incentives

10, 173–196.

promote successful ﬁsheries. Nature 470 (7334), 386–389.

Aloo, P.A., Munga, C.N., Kimani, E.N., Ndegwa, S., 2014. A review of the status and

Hardman, E.R., Edwards, A.J., Rafﬁn, J.S.J., 2013. The seine-net ﬁshery of Rodrigues

potential of the coastal and marine ﬁsheries resources in Kenya. Int. J. Mar. Sci.

4. Island western Indian Ocean: is it sustainable or in terminal decline? Fish. Res.

139, 35–42.

Arias-Gonzalez, J.E., Galzin, R., Nielson, J., Mahon, R., Aiken, K., 1994. Reference area

Harrison, P., Samoilys, M., 2009. Livelihood Opportunities for Coastal Communities

as a factor affecting potential yield estimates of coral reef ﬁshes. NAGA ICLARM

in Kiunga, Kenya: Prospects for Livelihood Diversiﬁcation Amid Improved

Q. 17, 37–40.

Natural Resources Management in Kiunga Marine National Reserve

Ateweberhan, M., McClanahan, T.R., Graham, N.A.J., Sheppard, C.R.C., 2011.

Kibodo/Kilimanyika/CORDIO Report, 58pp. Available from www.cordioea.net.

Episodic heterogeneous decline and recovery of coral cover in the Indian

Ocean. Coral Reefs 30, 739–752.

190 M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191

Henson, S., Winnie, M., 2004. Kenyan exports of Nile perch: the impact of food McClanahan, T.R., Muthiga, N.A., Kamukuru, A.T., Machano, H., Kiambo, R.W., 1999.

safety standards on an export-oriented supply chain. In: World Bank Policy The effects of marine parks and ﬁshing on coral reefs of northern Tanzania.

Research Working Paper 3349, June 2004. Biol. Conserv. 89, 161–182.

Hicks, C.C., McClanahan, T.R., 2012. Assessing gear modiﬁcations needed to McClanahan, T., Maina, J., Pet-Soede, L., 2002. Effects of the 1998 coral mortality

optimize yields in a heavily exploited, multi-species, seagrass and coral reef event on Kenyan coral reefs and ﬁsheries. Ambio 31, 543–550.

ﬁshery. PLoS One 7 (5), e36022. McClanahan, T.R., Mwaguni, S., Muthiga, N.A., 2005. Management of the Kenyan

Jennings, S., 1998. Artisanal ﬁsheries of the Great Astrolabe Reef, Fiji: monitoring, coast. Ocean Coast. Manag. 48, 901–931.

assessment and management. Coral Reefs 17, 82. McClanahan, T.R., Graham, N.A.J., Calnan, J.M., MacNeil, M.A., 2007. Toward

Jennings, S., Polunin, N.V.C., 1995. Comparative size and composition of yield from pristine biomass: reef ﬁsh recovery in coral reef marine protected areas in

six Fijian reef ﬁsheries. J. Fish Dis. 46, 28–46. Kenya. Ecol. Appl. 17 (4), 1055–1067.

Jennings, S., Polunin, N.V.C., 1996. Impacts of ﬁshing on tropical reef ecosystems. McClanahan, T.R., Hicks, C.C., Darling, E.S., 2008. Malthusian overﬁshing and efforts

Ambio 25 (1), 44–49. to overcome it on Kenyan coral reefs. Ecol. Appl. 18, 1516–1529.

Jennings, S., Kaiser, M.J., Reynolds, J.D., 2001. Marine Fisheries Ecology. Blackwell McClanahan, T.R., Kaunda-Arara, B., Omukoto, J.O., 2010. Composition and

Science Ltd., London. diversity of fish and fish catches in closures and open-access fisheries of Kenya.

Jennings, S., 2009. The role of marine protected areas in environmental Fish. Manag. Ecol. 17, 63–76.

management. ICES J. Mar. Sci. 66 (1), 16–21. McClanahan, T., Allison, E.H., Cinner, J.E., 2013. Managing ﬁsheries for human and

Karuga, S., Abila, R., 2007. Value Chain Market Assessment for Marine Fish food security. Fish Fish. 16 (1), 78–103.

Sub-sector in Kenya’s Coast Region. Market Economies Development Ltd., P.O. McClanahan, T.R., 1994. Kenyan coral reef lagoon ﬁsh: associations with reef

Box 1 Box 10423-00400 Nairobi. management, complexity, and sea urchins. Coral Reefs 13, 231–241.

Kaunda-Arara, B., Rose, G.A., 2004. Effects of marine reef National Parks on ﬁshery McGuiness, K.A., 2007. Effects of Beach Seine Fishery on Populations of

CPUE in coastal Kenya. Biol. Conserv. 118, 1–13. Seagrass-associated Leptoscarus Vaigiensis and Siganus Sutor. MSc Thesis.

Kaunda-Arara, B., Rose, G.A., Muchiri, M.S., Kaka, R., 2003. Long-term trends in Department of Biology. University of York, York, England.

coral reef ﬁsh yields and exploitation rates of commercial species from costal Mees, C.C., Yeeting, B.M., Taniera, T., 1988. Small scale ﬁsheries in the Gilbert group

Kenya. WIO J. Mar. Sci. 2, 105–116. of the Republic of Kiribati. Document OIREP1/2/3/4/C/R. Fisheries Division,

Kawaka, J., Samoilys, M.A., Church, J., Murunga, M., Abunge, C., Maina, G.W., 2015. Ministry of Natural Resource Development, Tarawa, Kiribati. 63 p.

Locally Managed Marine Areas (LMMAs) in Kenya: a Detailed History of Their Munga, C.N., Mohamed, M.O., Amiyo, N., Dahdouh-Guebas, F., Obura, D.O.,

Development and Establishment CORDIO/UNDP/SGP Report, 29pp. Available Vanreusel, A., 2012. Status of coral reef ﬁsh communities within the Mombasa

from www.cordioea.net. Marine Protected Area, Kenya, more than a decade after establishment. WIO J.

Koslow, J.A., Aiken, K., Auil, S., Clementson, A., 1994. Catch and effort analysis of Mar. Sci. 10 (2), 169–184.

the reef ﬁsheries of Jamaica and Belize. Fish. Bull. 92, 737–747. Munro, J.L., 1969. The sea ﬁsheries of Jamaica: past, present and future. Jamaica J. 3,

Kroeker, K.J., Kordas, R.L., Crim, R., Hendriks, I.E., Ramajo, L., Singh, G.S., et al., 2013. 16–22.

Impacts of ocean acidiﬁcation on marine organisms: quantifying sensitivities Munro, J.L., Williams, D. McB., 1985. Assessment and management of coral reef

and interaction with warming. Glob. Change Biol. 19 (6), 1884–1896. ﬁsheries: biological, environmental and socio-economic aspects. Tahiti, 27

Kuster, C., Vuki, V.C., Zann, L.P., 2005. Long term trends in subsistence ﬁshing May–1 June 1985 In: Proc. 5th Int. Coral Reef Congress, 4, pp. 543–578,

patterns and coral reef ﬁsheries yield from a remote Fijian village. Fish. Res. 76, Antenne Museum-EPHE, Moonea, French Polynesia.

221–228. Muthiga, N., Costa, A., Motta, H., Muhando, C., Mwaipopo, R., Schleyer, M., 2008.

Laroche, J., Ramananarivo, N., 1995. A preliminary survey of the artisanal ﬁshery Status of coral reefs in east africa: Kenya, Tanzania, Mozambique and South

on coral reefs of the Tulear Region (Southwest Madagascar). Coral Reefs 14 (4), Africa. In: Wilkinson, C. (Ed.), Status of Coral Reefs of the World. Global Coral

193–200. Reef Monitoring Network and Reef and Rainforest Research Center, Townsville,

Locham, G.A., Kaunda-Arara, B., Mlewa, C.M., 2010. The inﬂuence of reef type and Australia, 296pp.

seasonality on population structure of coral-reef ﬁshes within Malindi Marine Newton, K., Côté, I.M., Pilling, G.M., Jennings, S., Dulvy, N.K., 2007. Current and

Park, Kenya. Mar. Ecol. 31, 494–505. future sustainability of Island coral reef ﬁsheries. Curr. Biol. 17 (7), 655–658.

Luckhurst, B.E., Trott, T.M., 2009. Seasonally-closed spawning aggregation sites for Nzioka, R.M., 1990. Fish yield of Kiliﬁ coral reefs in Kenya. Hydrobiologia 208 (1–2),

Red Hind (Epinephelus guttatus): Bermuda’s experience over 30 years 81–84.

(1974–2003). Proceedings of the 61st Gulf and Caribbean Fisheries Institute, Obura, D.O., Muthiga, N.A., Watson, M., 2000. Kenya. In: McClanahan, T.R.,

331–336. Sheppard, C.R.C., Obura, D.O. (Eds.), Coral Reefs of the Indian Ocean: Their

Maina, G.W., Obura, D., Alidina, H., Munywoki, B., 2008. Increasing catch in an Ecology and Conservation. Oxford University Press, Inc., New York, New York,

overexploited reef ﬁshery − Diani-Chale, Kenya, from 1998 to 2006. In: Obura, pp. 199–229.

D.O., Tamelander, J., Linden, O. (Eds.), Ten Years After Bleaching −Facing the Obura, D.O., 2001. Participatory monitoring of shallow tropical marine ﬁsheries by

Consequences of Climate Change in the Indian Ocean. CORDIO Status Report artisanal ﬁshers in Diani, Kenya. Bull. Mar. Sci. 69, 777–792.

2008. , Available from www.cordioea.net. Ochiewo, J., 2004. Changing ﬁsheries practices and their socioeconomic

Maina, G.W., Samoilys, M., Alidina, H., Osuka, K., 2013. Targeted ﬁshing of the implications in South Coast Kenya. Ocean Coast. Manag. 47 (7-8), 389–408.

shoemaker spinefoot rabbitfish, Siganus sutor, on potential spawning Pauly, D., 1988. Some definitions of overfishing relevant to coastal zone

aggregation in southern Kenya. In: Robinson, J., Samoilys, M. (Eds.), Reef Fish management in Southeast Asia. Trop. Coast. Area Manag. 3, 14–15.

Spawning Aggregations in the Western Indian Ocean: Research for Perry, R.I., Cury, P., Brander, K., Jennings, S., Möllmann, C., Planque, B., 2010.

Management. WIOMSA/SIDA/SFA/CORDIO. WIOMSA Book Series. , p. 13. Sensitivity of marine systems to climate and ﬁshing: concepts, issues and

Mangi, S.C., Roberts, C.M., 2006. Quantifying the environmental impacts of artisanal management responses. J. Mar. Syst. 79, 427–435.

ﬁshing gear on Kenya’s coral reef ecosystems. Mar. Poll. Bull. 52, 1646–1660. Polunin, N.V.C., Roberts, C.M. (Eds.), 1996. Reef Fisheries. Chapman & Hall, London,

Mangi, S.C., Roberts, C.M., 2007. Factors inﬂuencing ﬁsh catch levels on Kenya’s 477p.

coral reefs. Fish. Manag. Ecol. 14, 245–253. Pomeroy, R., Parks, J., Mrakovcich, K.L., Lamonica, C., 2016. Drivers and impacts of

Mangi, S., 2006. Gear Management in Kenya’s Coastal Fisheries. Doctor of ﬁsheries scarcity, competition, and conﬂict on maritime security. Mar. Policy

Philosophy Thesis. University of York, Environment Department, UK. 67, 94–104.

Marriott, S.P., 1984. Rural ﬁsheries in Kiribati: Frame surveys of Abemama, Kuria Richards, L.J., Schnute, J.T., 1986. An experimental and statistical approach to the

and North Tarawa. Tarawa: Ministry of Natural Resources Development. 30 p. question: is CPUE an index of abundance? Can. J. Fish. Aquat. Sci. 43,

Marten, G.G., Polovina, J.J., 1982. A comparative study of ﬁsh yields from various 1214–1227.

tropical ecosystems. In: Pauly, D., Murphy, G.I. (Eds.), Theory and Management Roberts, C.M., Polunin, N.V., 1991. Are marine reserves effective in management of

of Tropical Fisheries. , pp. 255–289, ICLARM Conference Proceedings 9. reef ﬁsheries? Rev. Fish Biol. Fish. 1 (1), 65–91.

International Center for Living Aquatic Resources Management, Manila, Robinson, J., Samoilys, M.A., Grandcourt, E., Julie, D., Cedrasa, M., Gerry, G., 2011.

Philippines, and Division of Fisheries Research, CSIRO, Cronulla Australia. The importance of targeted spawning aggregation ﬁshing to the management

Maypa, A.P., Russ, G.R., Alcala, A.C., Calumpong, H.P., 2002. Long-term trends in of Seychelles’ trap ﬁshery. Fish. Res. 112, 96–103.

yield and catch rates of the coral reef ﬁshery at Apo Island: central Philippines. Rocliffe, S., Peabody, S., Samoilys, M., Hawkins, J.P., 2014. Towards a network of

Mar. Freshw. Res. 53, 1–7. locally managed marine areas (LMMAs) in the Western Indian Ocean. PLoS One

McClanahan, T.R., Hicks, C., 2011. Changes in life history and ecological 9 (7).

characteristics of coral reef fish catch composition with increasing fishery Russ, G.R., Alcala, A.C., 1989. Effects of intense fishing pressure on an assemblage of

management. Fish. Manag. Ecol. 18, 50–60. coral reef ﬁshes. Mar. Ecol. Prog. Ser. 56, 13–27.

McClanahan, T.R., Mangi, S., 2000. Spillover of exploitable ﬁshes from a marine Russ, G.R., Alcala, A.C., 1998. Natural ﬁshing experiments in marine reserves

park and its effect on the adjacent ﬁshery. Ecol. Appl. 10, 1792–1805. 1983–1993: roles of life history and ﬁshing intensity in family responses. Coral

McClanahan, T.R., Mangi, S., 2001. The effect of a closed area and beach seine Reefs 17, 399–416.

exclusion on coral reef ﬁsh catches. Fish. Manag. Ecol. 8, 107–121. Russ, G.R., 1991. Coral reef ﬁsheries: effects and yield. In: Sale, P.F. (Ed.), The

McClanahan, T.R., Mangi, S., 2004. Gear-based management of a tropical artisanal Ecology of Fishes on Coral Reefs. Academic Press, San Diego, CA, pp. 601–636.

ﬁshery based on species selectivity and capture size. Fish. Manag. Ecol. 11, Russ, G.R., 2002. Yet another review of marine reserves as reef ﬁshery management

52–60. tools. In: Sale, P.F. (Ed.), Coral Reef Fishes: Dynamics and Diversity in a

McClanahan, T.R., Obura, D., 1995. Status of Kenyan coral reefs. Coast. Manag. 23, Complex Ecosystem. Academic Press, San Diego, California USA, pp. 421–443.

57–76. Samoilys, M., Tuda, P.M., 2009. Capacity Building, Awareness and Consultation

Workshops and Exchange Visits for Artisanal Fisheries in Kenya CORDIO

Report to FAO, 40pp. Available from www.cordioea.net.

M.A. Samoilys et al. / Fisheries Research 186 (2017) 177–191 191

Samoilys, M.A., Maina, G.W., Osuka, K., 2011a. Artisanal ﬁshing gears of the kenyan Tuda, P., Nyaga, W., Maina, G.W., Wanyonyi, I., Obura, D.O., 2008. Estimating total

coast mombasa. CORDIO, 36. ﬁshing effort over tidal to annual periods in the Diani-Gazi reef ﬁshery in

−

Samoilys, M.A., Maina, G.W., Osuka, K., 2011b. Policy Brief October 2011: Kenya. In: Obura, D.O., Tamelander, J., Linden, O. (Eds.), Ten Years After

Recommendations for Sustainable and Responsible Fishing in Kenya’s Coastal Bleaching −Facing the Consequences of Climate Change in the Indian Ocean. ,

Artisanal Fisheries, 5pp. Available from www.cordioea.net. pp. 321–334, CORDIO Status Report 2008. Available from www.cordioea.net.

Samoilys, M., Kanyange, N., Macharia, D., Maina, G.W., Robinson, J., 2013. Dynamics Tummers, B., 2006. DataThief III, http://datathief.org/ (Accessed: 22.08.10).

of rabbitﬁsh (Siganus sutor) spawning aggregations in southern Kenya. In: Wass, R.C., 1982. The shorelines ﬁshery of American Samoa: past and present. In:

Robinson, J., Samoilys, M. (Eds.), Reef Fish Spawning Aggregations in the Marine and Coastal Processes in the Paciﬁc: Ecological Aspects of Coastal Zone

Western Indian Ocean: Research for Management. , WIOMSA Book Series 13. Management, J.L. Munro. UNESCO-ROSTEA, Jakarata, pp. 51–83.

Samoilys, M.A., 1988. Abundance and species richness of coral reef fish on the Watson, M., Ormond, R.F.G., 1994. Effect of an artisanal fishery on the fish and

Kenyan Coast: the effects of protective management and ﬁshing. Proc. 6th Int. urchin populations of a Kenyan coral reef. Mar. Ecol. Prog. Ser. 109, 115–129.

Coral Reef Symp 2, 261–266. Watson, M., Righton, D.A., Austin, T.J., Ormond, R.F.G., 1996. The effects of ﬁshing

Sheppard, C.R.C., Wells, S.M., 1988. Coral Reefs of the World, vol. 2, Indian Ocean, on coral reef ﬁsh abundance and diversity. JMBA 76, 229–233.

Red Sea and Gulf. UNEP Regional Seas Directories and Bibliographies. IUCN, Wheeler, J., Ommanney, F., 1953. Report on the Mauritius-Seychelles ﬁsheries

Gland, Switzerland and UNEP Nairobi, pp. i+ 389+36 maps. survey 1948–1949, part 4. Fish. Publ. Lond. 1 (3), 120–140.

Stevenson, D.K., Marshall, N., 1974. Generalisations on the ﬁsheries potential of Wright, A., Richards, A.H., 1985. A multispecies ﬁsheries associated with coral reefs

coral reefs and adjacent shallow-water environments. In: Proc. 2nd Int. Coral in the Tigak islands, Papua New Guinea. Asian Mar. Biol. 2, 69–84.

Reef Symp. Australia, June 1973, Great Barrier Reef Committee, Brisbane, pp. Zar, J.H., 1996. Biostatistical Analysis, 3rd ed. Prentice-Hall Inc., New Jersey, 662 pp. 147–156.