Estimating hunting prevalence and reliance on wild meat in Cambodia's Eastern Plains

Page created by Nathan Wong
 
CONTINUE READING
Estimating hunting prevalence and reliance on wild meat in Cambodia's Eastern Plains
Estimating hunting prevalence and reliance on wild
            meat in Cambodia's Eastern Plains
                                   HARRIET IBBETT, AIDAN KEANE, ANDREW D.M. DOBSON, OLLY GRIFFIN
                                                            H E N R Y T R A V E R S and E . J . M I L N E R - G U L L A N D

            Abstract Hunting is a primary driver of biodiversity loss                                   Introduction
            across South-east Asia. Within Cambodia, the use of wire
                                                                                                                unting threatens % of terrestrial mammal species
            snares to capture wildlife is a severe threat in protected
            areas but there have been few studies of the behaviour of                                   H       globally (Ripple et al., ), and is estimated to
                                                                                                        have decreased bird and mammal abundances by  and
            hunters from local communities. Here, we combine the
            unmatched count technique with direct questioning to                                        %, respectively, in some tropical areas (Benitez-Lopez
            estimate the prevalence of hunting behaviours and wildlife                                  et al., ). The situation is particularly severe in South-
            consumption amongst  households living within Keo                                        east Asia, where most large wild vertebrate species have
            Seima Wildlife Sanctuary, Cambodia. We assessed respon-                                     declined substantially throughout their remaining ranges
            dents’ knowledge of rules, and their perceptions of patrols                                 (Sodhi et al., ; Harrison et al., ). Here, forests are
            responsible for enforcing rules. Estimates of hunting be-                                   increasingly considered empty; i.e. devoid of all but the
            haviour were variable: results from the unmatched count                                     smallest or most common species (Harrison, ), with
            technique were inconclusive, and direct questioning re-                                     overexploitation facilitated by advancements in hunting
            vealed % of households hunted, and % set snares around                                   technologies, rapid economic growth and improved access
            farms to prevent wildlife eating crops. Hunting with domes-                                 to forested areas (Harrison et al., ; Hughes, ).
            tic dogs was the method most commonly used to catch                                             Cambodia is one of the most biodiverse countries in
            wildlife (% of households owned dogs). Wild meat was                                      South-east Asia (Daltry, ) and is legally one of the
            consumed by % of households, and was most frequently                                      best protected, with % of terrestrial land area afforded
            bought or caught, but also gifted. We detected a high aware-                                protected status (Souter et al., ). However, in reality pro-
            ness of conservation rules, but low awareness of punish-                                    tected areas are chronically underfunded, overexploitation
            ments and penalties, with wildlife depletion, rather than                                   of natural resources is widespread, and laws are weakly en-
            the risk of being caught by patrols, causing the greatest                                   forced (Souter et al., ). Hunting has probably driven
            reduction in hunting. Our findings demonstrate the chal-                                    the kouprey Bos sauveli to extinction, extirpated the Javan
            lenges associated with reliably estimating rule-breaking                                    rhinoceros Rhinoceros sondaicus and tiger Panthera tigris
            behaviour and highlight the need to incorporate careful                                     (O’Kelly et al., ), and continues to threaten the viability
            triangulation into study design.                                                            of many other species (Starr et al., ; O’Kelly et al., ;
                                                                                                        Rostro-García et al., ). Cambodia has a high prevalence
            Keywords Bushmeat, Cambodia, law enforcement, poach-                                        of hunting and reliance on wild meat, with an estimated %
            ing, protected areas, ranger patrols, snares, unmatched                                     of rural households engaged in some form of harvest of wild
            count technique                                                                             animals at least once per year (Nielsen et al., ). Snares
            Supplementary material for this article is available at                                     are widely used; in , , were removed by patrols in
            doi.org/./S                                                           Cambodia’s South Cardamoms National Park (Gray et al.,
                                                                                                        ), with true snare abundance probably much higher,
                                                                                                        as experimental studies suggest only a small proportion of
                                                                                                        snares set are found by rangers (O’Kelly et al., ; Ibbett
                                                                                                        et al., ). Usually made from wire, cable or nylon, snares
                                                                                                        are affordable, accessible, and have limited selectivity with
                                                                                                        respect to animals’ species, sex or age (Noss, ). Once
                                                                                                        set, they can trap a wide range of arboreal and terrestrial
            HARRIET IBBETT (Corresponding author,     orcid.org/0000-0003-1213-4834),                   species (Borgerson, ; Ingram et al., ) and, although
            HENRY TRAVERS and E. J. MILNER-GULLAND Department of Zoology, University                    animals occasionally escape, subsequent non-fatal injuries
            of Oxford, Oxford, OX1 3SZ, UK. E-mail harriet.ibbett@bangor.ac.uk
                                                                                                        often jeopardize long-term survival (Yersin et al., ).
            AIDAN KEANE and ANDREW D. M. DOBSON School of Geosciences, University of
            Edinburgh, Edinburgh, UK
                                                                                                            The impact of hunting on Cambodia’s fauna is well docu-
                                                                                                        mented (Harrison et al., ), but there is less empirical
            OLLY GRIFFIN Wildlife Conservation Society—Cambodia Programme, Phnom
            Penh, Cambodia                                                                              information about hunters, their hunting methods, and
            Received  September . Revision requested  November .                            local demand for wildlife products (but see Martin &
            Accepted  December .                                                                   Phipps, ; Loucks et al., ; Coad et al., ). This

              This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives licence (http://creativecommons.org/licenses/by-nc-nd/4.0/),
              which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is unaltered and is properly cited. The written permission of Cambridge
DownloadedUniversity
               from https://www.cambridge.org/core.         IP address:
                          Press must be obtained for commercial    re-use46.4.80.155,  oncreate
                                                                          or in order to  19 Dec  2020 at 07:37:34,
                                                                                                a derivative work. subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
              Oryx, Page 1 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
2         H. Ibbett et al.

                                                                                                                                             FIG. 1 Keo Seima Wildlife
                                                                                                                                             Sanctuary in Mondulkiri
                                                                                                                                             Province, Cambodia,
                                                                                                                                             indicating all settlements
                                                                                                                                             located within and close to the
                                                                                                                                             protected area (the  study
                                                                                                                                             villages are not identified, to
                                                                                                                                             ensure anonymity), and ranger
                                                                                                                                             patrol stations.

             may be partly because gathering robust information about                              semi-evergreen and evergreen forest in Mondulkiri and
             hunting is challenging, especially in contexts where it is a re-                      Kratie provinces on the eastern border of Cambodia (Fig. ;
             stricted or prohibited activity (Nuno & St John, ). Hunting                       Evans et al., ). The Sanctuary lies within a wider forested
             in Cambodian protected areas is a punishable offence                                  landscape known as the Eastern Plains, and supports re-
             (Forestry Administration, ; MoE, ), and thus those                            gionally important populations of Asian elephant Elephas
             who violate rules may not wish to identify themselves for                             maximus, wild cattle Bos spp., and globally important popu-
             fear of sanctions (Solomon et al., ). When asked directly,                        lations of primates (Nomascus gabriellae, Pygathrix nigripes;
             respondents may refuse to participate, provide inaccurate re-                         Griffin, ).
             sponses, conceal their true attitudes, beliefs or behaviours, or                          Approximately , people live within and around the
             temper their answers so as to appear more socially acceptable                         Sanctuary, the majority of whom are Bunong, an animist
             (known as ‘social desirability bias’; Tourangeau & Yan, ;                         minority Indigenous people who have strong spiritual con-
             Krumpal, ). Acquiring robust and reliable data on hunting                         nections to the forest and its wildlife. Traditionally, the
             prevalence is nonetheless important to ensure conservation                            Bunong practised swidden agriculture, and relied heavily on
             interventions are targeted towards the most appropriate                               non-timber forest products (NTFPs) such as honey, fish, rat-
             groups (St John et al., ; Jones et al., ).                                    tan, wild fruit, vegetables and wildlife for subsistence (Evans
                 Here, we quantify the prevalence of hunting amongst rural                         et al., ). The construction of roads has brought market
             communities in a Cambodian protected area. We use the un-                             integration to previously inaccessible villages and strength-
             matched count technique, an indirect questioning approach,                            ened cross-border trade links with Viet Nam (Mahanty &
             to estimate the prevalence of hunting as a subsistence and                            Milne, ). Many households have abandoned traditional
             income-generating livelihood activity. Use of a specialized                           swidden agricultural practices in favour of more profitable
             questioning technique can allow the biases typically asso-                            cash crops such as cassava and cashew (Travers et al., ).
             ciated with direct questioning to be overcome by assuring                             Villages, particularly those on the periphery, have experi-
             greater levels of anonymity, although often at the cost of                            enced large influxes of Khmer (the majority ethnic group in
             lower precision (Nuno & St John, ). We couple this                                Cambodia) families seeking land, and forest cover has de-
             with direct questioning to derive further information about                           clined as a result of subsequent clearance for small-scale
             seasonality, methods used, species caught and consumed,                               farms (Mahanty & Milne, ; Riggs et al., ). In addition,
             and trends in hunting activity. Finally, we assess local peoples’                     the forest has experienced severe pressure from illegal logging
             knowledge of rules regarding the capture and use of wildlife,                         for luxury timber, as well as industrial-scale forest clearance
             and their perceptions of the ranger patrols responsible for                           associated with government granting of Economic Land
             enforcing protected area rules.                                                       Concessions within protected area boundaries.
                                                                                                       Prior to , Keo Seima Wildlife Sanctuary was
                                                                                                   managed as a Protection Forest by the Government’s
             Study area                                                                            Forestry Administration. In  jurisdictional reforms of
                                                                                                   natural resource management resulted in transfer to the
             Our study was conducted in Keo Seima Wildlife Sanctuary,                              Ministry of Environment under sub-decree , and reclassi-
             a , km area of protected mixed deciduous dipterocarp,                            fication as a Wildlife Sanctuary, with the principal objective

                                                                                                Oryx, Page 2 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
Estimating hunting in Cambodia                     3

           of preserving and protecting wildlife (MoE, ). Accord-                            asked about any problems with wildlife on farms, and their
           ing to the  Forestry Law, it is strictly prohibited to                            perceptions of any change in hunting levels over the pre-
           hunt, harm or harass all wildlife, and under the                                  vious  years.
           Protected Area Law, killing wildlife and releasing hunting
           dogs is strictly forbidden, and regulated extraction of NTFPs
           and sustainable use is allowed in some zones within the                               Measuring hunting prevalence
           protected area (MoE, ). Hunting occurs throughout                                 To reduce social desirability and non-response biases, we
           the Sanctuary, however, and previous research has high-                               used the unmatched count technique to investigate the pro-
           lighted the reliance of local communities on wild meat for                            portion of households that collected wild meat, took snares to
           subsistence purposes (Travers, ) and to supplement in-                            the forest to hunt, and hunted to generate income in the pre-
           come. Hunting is also undertaken by outsiders (i.e. people                            vious year. Half the sample were randomly allocated to a con-
           who do not live within or immediately around the forest)                              trol group who received a list of non-sensitive items, and the
           for sport, commercial purposes and subsistence (Drury,                                treatment group received a list that included the same non-
           ; Evans et al., ). Responsibility for enforcing rules                         sensitive items, plus an additional sensitive item (Nuno &
           lies with – government rangers working across  pa-                              St John, ). Respondents were asked to report only the
           trol stations. Since  the government has received fi-                             number of items applicable to them, never which items.
           nancial and technical support for the management of the                               Item scores were averaged across groups, and the prevalence
           Sanctuary from the Wildlife Conservation Society.                                     of the sensitive item was estimated from the difference
                                                                                                 between the means. The unmatched count technique re-
                                                                                                 quires large sample sizes, and estimates can have wide stan-
           Methods
                                                                                                 dard errors (Hinsley et al., ); to mitigate against this, we
                                                                                                 employed a double-list unmatched count technique, in which
           Ethical considerations
                                                                                                 participants simultaneously act as control and treatment
           Upon arrival in each village, we met the village chief to explain                     groups by answering a control and treatment list for the
           our research aims and seek permission to work in the commu-                           same question, but with the set of non-sensitive items differ-
           nity. Before each interview the research purpose, risks, benefits,                    ing between the two lists (Droitcour et al., ). Prevalence of
           and proposed use of the data were explained to participants                           the sensitive item is derived by calculating the mean score
           before verbal consent was sought (Supplementary Material ).                          across the paired lists (Glynn, ). Because of high levels
           All interviews were voluntary, anonymous, and conducted in                            of illiteracy, pictorial lists were used, and items were verbally
           Khmer or Bunong by independent enumerators unassociated                               described to participants. A practice question on fruit con-
           with the Wildlife Conservation Society. Hunting is illegal, and                       sumption was used to introduce the method, and follow-up
           therefore to protect participants against reprisals questions on                      questions were asked to assess respondents’ understanding.
           hunting were targeted at the household rather than individual                         At the end of the interview, respondents were asked directly
           level, and village names have been anonymized here to offer                           whether they currently or had ever hunted and, if so, the spe-
           additional protection. All methods were piloted before data                           cies they caught, how often they hunted, the methods used,
           collection, in  households during February .                                    and their reasons for hunting.

           Household surveys                                                                     Knowledge of conservation rules

           During February–April , we interviewed people in                               We assessed knowledge of rules pertaining to hunting activity,
           households in  villages (Fig. ). Between  and % of                             and the perceived likelihood of () a neighbour knowing if
           households were surveyed per village, with houses identi-                             someone had caught wildlife, () being caught by a patrol
           fied using a systematic sampling strategy in which ques-                              when hunting and () receiving a penalty if caught. We mea-
           tionnaires were administered at every nth house, with n                               sured social acceptability by asking respondents whether they
           inversely related to village size. We interviewed any available                       would approve if a friend or family member went hunting.
           respondent above the age of  in each household. If respon-                          Finally, we asked households whether they had ever been caught
           dents declined or were absent, interviews were conducted                              by a patrol in possession of wildlife, and if so what happened.
           at the next available house. Enumerators collected data on
           respondent demographics and household livelihood strat-                               Analysis
           egies, household reliance on wildlife species for meat and
           medicine, specifically the frequency with which species were                          Prior to analysis, unmatched count technique data were
           consumed, how wildlife was accessed, and the most preferred                           tested to determine whether individual responses to
           meat (Supplementary Material  & ). Respondents were also                            the non-sensitive item changed depending upon the

           Oryx, Page 3 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
4         H. Ibbett et al.

             respondents’ treatment status (design effects). This test was
             conducted using the ict.test function in the list package
             of R .. (Blair et al., ; R Core Team, ). If a
             Bonferroni-corrected P-value of , . was detected, this
             was interpreted as evidence for the presence of design
             effects, which were detected for one list (Supplementary
             Material ). In addition, both floor and ceiling effects were
             detected for all lists. Floor effects occur if a higher than ex-
             pected number of respondents give an aswer of  (i.e. none
             of the items were applicable), and ceiling effects occur if the
             maximum number of items is reported for each list (i.e. all
             the items were applicable). Both effects reduce anonymity
             and indicate the method did not function as expected.
             Prevalence estimates were calculated by combining the
             scores from list pairs using the ictreg function in R.                                FIG. 2 Temporal change in the prevalence of hunting reported
                                                                                                   by  respondents in Keo Seima Wildlife Sanctuary in .

             Results

             In total, % of respondents were men, % were women.
             Respondents were Bunong (%), Khmer (%) or from
             other Indigenous groups (Stieng, Laotian, Cham; %).
             Respondents had lived in their village for a mean of  ±
             SD . years, with  ± SD . years of formal education.
             The main sources of household income were farming
             (% of households), shops or businesses (%), resin col-
             lection (%), opportunistic paid labour (%), salaried work
             (%) or illegal logging (%). Eighty-four per cent of house-
             holds collected NTFPs, and % collected resin.

                                                                                                   FIG. 3 Triangulated estimates of hunting prevalence in Keo
             Wildlife hunting                                                                      Seima Wildlife Sanctuary, with % confidence intervals.
                                                                                                   DQ indicates questions in which  respondents where asked
             Hunting prevalence, frequency & seasonality                                           directly about hunting. UCT indicates findings from the
             When directly questioned, % of households reported hunt-                             unmatched count technique ( of  respondents answered
             ing, and % of households reported they used to hunt but                             questions).
             no longer did. When asked to indicate the year they ceased
             hunting (Fig. ), % of hunters indicated they stopped
             hunting after . Reasons included increased difficulty                             who reported hunting (, %; Fig. ). Unmatched count
             in catching wildlife (% of retired hunters), reduced time                           technique estimates for taking snares to the forest and hunt-
             available for hunting (%), lack of dogs to hunt with                                ing for income were also unreliable and did not significantly
             (%), old age (%) and concern about meeting law en-                                  differ from zero (Supplementary Material ).
             forcement patrols (%). The unmatched count technique                                     In households where people directly reported hunting
             warm-up question regarding prevalence of fruit consump-                               (% of all respondents) there was a mean of . hunting trips
             tion appeared to work as expected, estimating  ± SE %                              per month, almost double the number reported by retired
             of respondents consumed the sensitive fruit, but the ques-                            hunters (% of respondents, . trips per month). When
             tion on hunting provided a negative prevalence estimate                               asked what they would do if hunting became harder (i.e. wild-
             (Fig. ). Ideally, no respondent should report that all items                         life was caught less frequently), % of current hunters would
             on either the control or treatment list were applicable to                            stop, % would seek new hunting grounds, % would continue
             them (i.e. nobody should answer  or ), but responses to                             to collect other NTFPs (hunting if the opportunity arose),
             the question were subject to ceiling effects, meaning some                            % would change method, and % were unsure. Seventy-
             respondents reported all items on the lists were applicable.                          seven per cent of all respondents thought more hunting
             Although this undermines assurances of anonymity (be-                                 occurred in the wet season when the absence of leaf litter
             cause the interviewer knows the sensitive answer applies                              on the forest floor made it easier to walk quietly in the forest,
             to the respondent), it provides a direct count of households                          snares could be set around fruiting trees such as wild almond

                                                                                                Oryx, Page 4 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
Estimating hunting in Cambodia                     5

           Irvingia malayana, and wildlife was easier to catch as animals                        TABLE 1 Hunting methods reported by a total of  retired hunters
           were distracted while foraging on new growth. In addition, it                         (% of respondents surveyed) and  current hunters (% of re-
           was reported that lulls in agriculture and logging meant peo-                         spondents surveyed) in Keo Seima Wildlife Sanctuary, Cambodia,
                                                                                                 in .
           ple had more time to allocate to hunting, and poor road con-
           ditions reduced the chance of encountering ranger patrols.                                                                  No. of retired          No. of current
           Six per cent of respondents indicated hunting was more fre-                           Method                                hunters (%)             hunters (%)
           quent in the dry season when the absence of foliage made it                           Dogs                                  155 (81)                52 (87)
           easier for dogs to run unhindered, and when water scarcity                            Slingshot                              73 (38)                28 (47)
           meant efforts could be targeted around water sources. Sev-                            Snares                                 40 (21)                 8 (13)
           enteen per cent of respondents did not know when most                                 Crossbow                               12 (6)                  2 (3)
           hunting occurred.                                                                     Gun                                     5 (3)                  0 (0)
                                                                                                 Total no. of respondents               192                     60

           Hunting methods
           Dogs were the most commonly reported hunting method                                   Protecting crops
           (% of  current hunters), followed by slingshot (%),                             Alongside killing wildlife for food, medicine and income,
           snares (%) and crossbow (%; Table ). Only % of all                               respondents reported killing wildlife to protect crops. Sev-
           respondents reported ever having set snares in the forest,                            enty-one per cent of all households reported that wildlife
           and only eight interviewees reported currently doing so                               ate or destroyed crops, and % reported setting snares (a
           (%), setting a mean of  snares, although one respondent                            mean of  per household) around farms to protect crops.
           reported maintaining  snares.                                                      Four per cent of households reported setting –
               People in % of current hunting households reported                              snares. The main problem species reported were the wild
           using more than one method to catch wildlife. Dogs                                    boar (% of respondents with wildlife problems), long-tailed
           and slingshots were reported more frequently by current                               macaque (%), elephant (%) and green peafowl Pavo
           hunters than retired hunters, with snares, crossbows and                              muticus (%). Other species mentioned included the East
           guns reported less frequently by current hunters (Table ).                           Asian porcupine Hystrix brachyura, red muntjac, jungle
           Respondents often said that guns were only used by people                             fowl and bamboo rat (Rhizomyini spp.).
           outside the community or authorities, such as police or the
           military. Several respondents reported seeing soldiers hunting                        Hunting by dogs
           primates, with guns, and another stated they had seen men
                                                                                                 Seventy-nine per cent of households owned dogs, and % of
           with rifles, in  ×  vehicles, with cool-boxes to take meat
                                                                                                 those who went to the forest took dogs with them for com-
           away. One respondent said they had borrowed a gun from
                                                                                                 panionship and for protection against encounters with wild-
           the police to shoot black-shanked douc langur P. nigripes.
                                                                                                 life. Although a few respondents reported actively using dogs
               Of those that reported hunting, % said they did so only
                                                                                                 to hunt species such as muntjac and sambar, many of those
           for subsistence, % hunted for food and income, and only
                                                                                                 who reported dogs killing wildlife said it was unintentional.
           one household reported hunting solely for income. One indi-
                                                                                                 When accompanying owners to the forest, dogs would
           vidual said that, if successful, they could earn USD –
                                                                                                 chase wildlife scents. Owners also reported dogs roaming
           per month selling meat to villagers or external traders. This
                                                                                                 away from home to hunt, catching species such as monitor
           is considerably more than the monthly earnings of a casual
                                                                                                 lizards, chevrotains, turtles and tortoises. Excluding puppies,
           labourer, which are KHR ,–, per day (USD –
                                                                                                 we recorded a total of , dogs; the mean number of dogs
            per month). The most commonly caught species were
                                                                                                 owned across all  households was . (a mean of .
           monitor lizard (Varanus spp., % of current hunters), wild
                                                                                                 dogs per  dog-owning households). Approximately %
           boar Sus scrofa (%), chevrotain (Tragulus spp., %) and
                                                                                                 of households were surveyed, suggesting the total number
           civets (Paradoxurus hermaphroditus, Viverra zibetha, %;
                                                                                                 of domestic dogs living within Keo Seima Wildlife Sanc-
           Table ). Other species caught included northern red munt-
                                                                                                 tuary could exceed ,, or . per km, a higher density
           jac Muntiacus vaginalis, sambar Rusa unicolor, long and pig-
                                                                                                 than many species of conservation interest (Wildlife Conser-
           tailed macaque (Macaca fascicularis, Macaca nemestrina), black-
                                                                                                 vation Society, unpubl. data).
           shanked douc langur, southern yellow-cheeked crested gibbon
           Nomascus gabriellae, Sunda colugo Galeopterus variegatus,
           red jungle fowl Gallus gallus, and various tortoise, turtle and
                                                                                                 Wildlife use and consumption
           squirrel species. Compared to retired hunters, a greater
           proportion of current hunters reported catching monitor                               Eighty-five per cent of all households consumed wild meat,
           lizards and civets, and fewer reported catching wild boar,                            and % used wildlife products for medicinal purposes.
           muntjacs and chevrotains (Table ).                                                   Overall, % of respondents preferred eating wild meat

           Oryx, Page 5 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
6         H. Ibbett et al.

             TABLE 2 Species most commonly caught, as reported by a total of  current and retired hunters in Keo Seima Wildlife Sanctuary,
             Cambodia, in .

             Species                                                 National status (Red List status1) No. of retired hunters (%) No. of current hunters (%)
             Monitor lizard Varanus spp.                             Common (LC)                                116 (60)                           43 (71)
             Wild boar Sus scrofa                                    Common (LC)                                 78 (41)                           17 (28)
             Chevrotain Tragulus spp.                                Unclassified (LC)                           46 (24)                            8 (13)
             Northern red muntjac Muntiacus vaginalis                Common (LC)                                 26 (15)                            2 (3)
             Civets2                                                 Common (LC)                                  9 (5)                             7 (12)
             Primates3                                               Common/Rare (LC/EN)                         11 (6)                             3 (6)
             Sambar Rusa unicolor                                    Common (VU)                                  2 (1)                             1 (, 1)
             Total no. of repondents                                                                             192                                60
             
              LC, Least concern; VN, Vulnerable; EN, Endangered.
             
               Common palm civet Paradoxurus hermaphroditus and large Indian civet Viverra zibetha.
             
               Long-tailed/pig-tailed macaque Macaca fascicularis/nemestrina: Common/LC; black-shanked douc langur Pygathrix nigripes and yellow-cheeked crested
             gibbon Nomascus gabriellae: Rare/EN.

             TABLE 3 Reasons given by  respondents living in Keo Seima                          % reported being given wild meat by family or neigh-
             Wildlife Sanctuary when asked about their preference for differ-                      bours. More households ate wild meat in the wet (% of
             ent types of meat in  (six respondents indicated they had no                      consumers) than in the dry season (%), with wild meat
             preference, and six did not answer the question).
                                                                                                   also consumed more frequently in the wet (mean .
                                                Meat type preferred                                times per month) than in the dry season (mean . times
             Reason for preference              Wild meat (%)         Domestic meat (%)
                                                                                                   per month). This corroborates the reportedly higher fre-
                                                                                                   quency of wildlife interactions in the wet season, when
             It is better for your health       299 (42)              16 (2)
             It has no chemicals                235 (33)              5 (, 1)
                                                                                                   species can be caught eating crops. In addition, domestic
             It is tastier                      209 (30)              21 (3)                       meat alternatives were reportedly less available during the
             It is natural                      58 (8)                –                            wet season, as rain restricts traders’ access and villagers
             It is more affordable              1 (, 1)               7 (, 1)                      have less income to purchase domestic meat.
             It is easier to buy                –                     122 (17)                         The species most commonly reported as eaten were wild
             Total no. of respondents           490 (70)              166 (24)                     boar (%), monitor lizard (%), muntjac (%), chevrotain
                                                                                                   (%), monkeys (%), civets (%), sambar (%), tortoises and
                                                                                                   turtles (%) and jungle fowl (%). Snakes, porcupine and
             to domestic alternatives, mostly because wild meat was                                other rodents accounted for , % each. Species most likely
             believed to be healthier, free from chemicals, and chnganh                            to be bought were wild boar, monitor lizard and red muntjac.
             (delicious) (Table ). Twenty-four per cent of respondents                            Species most likely to be caught were monitor lizard, wild
             preferred domestic meat, mostly because it was more access-                           boar, chevrotain and muntjac, and monitor lizard was most
             ible. Domestic meat was reportedly more widely sold, and                              likely to be gifted (Table ). The most common species
             easier to buy in small quantities, than wild meat, which                              used for medicine were the slow loris Nycticebus pygmaeus
             was usually only sold by the kilogram. Several respondents                            (% of households) and the porcupine (%). In some villages,
             said wild meat could unknowingly be bought, as it is difficult                        slow loris could be ordered from local hunters, who caught
             to differentiate meat once butchered. Only % of respon-                              them at night using spotlights and slingshots. Others bought
             dents who preferred domestic meat cited affordability, and                            slow loris or porcupine from neighbours or traders, when
             prices often overlapped. Wild boar, for example, was USD                              available. One respondent reported that a tonic made from
             .–. per kg in villages, whereas domestic pork in the                            rice wine and slow loris, traditionally given to mothers
             district town (which is usually cheaper for commodities in                            after childbirth, could be purchased at one of the provincial
             general than villages) was USD –./kg. Muntjac and sam-                             markets. Other species mentioned more than once for medic-
             bar meat was less available, and more expensive than wild                             inal purposes included chevrotain (% of households), cobra
             boar meat (village price USD –./kg).                                              (Naja spp., %), flying squirrel (Pteromyini, %), muntjac
                Respondents reported accessing wild meat in several                                (%), civet (%), black-shanked douc (%), Sunda pangolin
             ways. It was most commonly (% of respondents) bought                                Manis javanica (%) and hornbill (Bucerotidae spp., , %).
             from villagers or motorcycle traders from the district town.                              The majority of respondents believed that since 
             Some said traders hid meat in compartments under their                                levels of hunting (% of respondents), wildlife consump-
             motorbike seats. Thirty-three per cent of respondents                                 tion (%) and sale of wildlife by villagers (%) had
             reported catching wild animals to eat, which is surprising                            decreased. Fifty-four per cent of respondents believed hunt-
             considering only a small proportion (%) reported hunting;                            ing levels had declined because wildlife was scarcer and

                                                                                                Oryx, Page 6 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
Estimating hunting in Cambodia                     7

           TABLE 4 χ tests of association between whether a species was re-
           ported as consumed by a household, and how wild meat was ac-
           cessed (with  degree of freedom). All species reported as eaten
           were tested, but only species for which there were positive asso-
           ciations with access type are reported.

           How household               Species household
           accessed wild meat          reported consuming            χ2           P
           Bought                      Wild boar                      58.266      , 0.001
                                       Monitor lizard                 19.338      , 0.001
                                       Muntjac                        18.362      , 0.001
           Caught                      Monitor lizard                121.020      , 0.001
                                       Wild boar                      40.765      , 0.001
                                       Chevrotain                     20.410      , 0.001        FIG. 4 The perception of  respondents regarding whether
                                       Muntjac                         7.271        0.007        neighbours would know about a villager’s hunting activity,
           Gifted                      Monitor lizard                  4.941        0.026        a villager being caught by a patrol if hunting, and a villager
                                                                                                 receiving a penalty if caught by a patrol in Keo Siema Wildlife
                                                                                                 Sanctuary.

           therefore harder to catch, % thought patrols deterred
           people, % said forest loss meant there was nowhere to
           hunt, and % said livelihood changes meant people were                                animals, such as elephants, was not. Twelve per cent of respon-
           now too busy farming cash crops to hunt. Reasons freely                               dents thought others would approve of hunting, and % did
           given by respondents for these changes included growing                               not know or had no opinion. If someone in the village caught
           village populations increasing demand for wildlife, and                               wildlife, % of respondents thought it likely that neighbours
           because hunters secretly sold wildlife to traders. Others                             would know (Fig. ). Some respondents said it was difficult
           suggested that increasing village populations meant people                            to keep it secret because children would spread the news,
           were less inclined to share wild meat with neighbours, to                             although secrecy would allow people to avoid sharing their
           avoid having to share wild meat with many people. Fifty-                              catch and reduce the risk of being reported.
           two per cent of respondents believed that hunting by
           outsiders had decreased, % of respondents did not know                              Perceptions of law enforcement effectiveness
           or thought hunting was not undertaken by outsiders, and                               If a villager hunted, only % of respondents thought it like-
           % thought hunting by outsiders had increased. Outsiders                              ly that a patrol would catch them, but if caught, % thought
           were typically regarded as people from outside the com-                               it likely a hunter would receive a penalty (Fig. ). Expected
           mune. Some respondents stated that declines in hunting                                penalties listed by respondents included arrest (% of re-
           by local people meant there was more wildlife, which                                  spondents), warning (%), fine (%), and confiscation of
           attracted outsiders to hunt.                                                          meat and/or snares (%). Twenty-four per cent of respon-
                                                                                                 dents did not know what the penalty would be. Respondents
                                                                                                 often stated that the type of penalty received depended on
           Law enforcement                                                                       the severity of the crime, and whether the hunter had pre-
                                                                                                 viously been caught. Some reported that if they were only
           Knowledge of rules                                                                    hunting for food, and had caught only small animals such
           Seventy-one per cent of respondents stated there were rules                           as wild boar, monitor lizards or tortoises, patrols may show
           about catching wildlife. Of these, % attributed their knowl-                        leniency. However, if caught hunting large animals such as
           edge of rules to the Wildlife Conservation Society (e.g. ‘WCS                         elephants, sambar or gaur, punishment could be a fine of
           said we cannot catch wildlife’). Twenty-seven per cent did not                        up to USD , or imprisonment.
           know if there were rules, and % believed there were none.                                Despite respondents saying there was a % probability
           When asked specifically about setting snares around farms,                            of being caught when hunting, overall, just  respondents
           % of respondents incorrectly believed it was legal, %                             (% of all those who reported ever hunting) had been caught
           correctly said it was not and % did not know.                                       by a patrol when hunting, and only once did a household
              The majority of respondents (%) thought that if a                                report severe punishment. In this incident, the respondent’s
           member of their household went hunting, their friends                                 son had been lent a gun by the police to shoot sambar. After
           and/or family would disapprove. However, some respondents                             his arrest by rangers, the police reportedly intervened and,
           explained that it depended on what was caught; using dogs                             rather than being prosecuted, the son was released with a
           to catch small animals, such as monitor lizards or turtles,                           fine of USD . Mostly, respondents reported receiving
           for food, was considered acceptable, whereas shooting large                           warnings or having meat confiscated.

           Oryx, Page 7 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
8         H. Ibbett et al.

             Perceptions of patrols                                                                uncertainty regarding prevalence. Direct questioning and
             Sixty-two per cent of respondents reported they were not                              ceiling counts from the unmatched count technique suggest
             worried about encountering patrols in the forest. The major-                          a prevalence of %, but the estimate from the latter did not
             ity said that as they did not hunt or partake in any illegal                          differ significantly from zero. This is probably the result of
             activities they had nothing to fear, although a small number                          floor and ceiling effects, which reduce precision, as well
             of respondents expressed concern that patrols might pre-                              as anonymity (Blair et al., ). Our findings also highlight
             vent or punish the legal collection of NTFPs, such as rattan.                         ambiguity regarding the definition of hunting; one-fifth of
             Others reported adapting their behaviour to avoid patrols, for                        respondents reported setting snares around farms, which
             example by waiting until patrols had passed. Some said                                was considered a legitimate crop protection activity, and
             friends and family would call to warn them if patrols passed                          when people were asked about how they accessed wild
             through the village towards the forest. One respondent said                           meat to eat, nearly a third said they caught their own wild
             when hunting in a group, each would travel individually to                            meat, presumably from snares set around farms. Yet, few
             reduce the chance of being spotted by a patrol, and meat                              households stated they hunted. Our results suggest that
             would be shared in the forest before leaving, to reduce pun-                          questions about the intentional killing of wildlife in the for-
             ishment if caught. In total, % of respondents expressed con-                        est were probably subject to bias, and that responses about
             cern about meeting patrols, of which % said that this was                           wildlife killed opportunistically (e.g. by dogs) or coinciden-
             because a member of their household was engaged in illegal                            tally (e.g. to protect crops) were less likely to be censored,
             logging. A further % of respondents said they were worried                          a matter also documented elsewhere in Cambodia (Coad
             that rangers would punish them if their dogs caught wildlife.                         et al., ). Our findings reinforce the need to consider
             Several respondents believed that camera traps set by the                             survey questions carefully, and to triangulate by asking
             Wildlife Conservation Society to monitor wildlife popula-                             several questions, in different ways, about the same topic,
             tions were actually set to photograph people hunting.                                 particularly if examining sensitive topics. During our
                                                                                                   research, some respondents were hushed by fellow family
                                                                                                   members when discussing hunting, and others failed to
             Corruption                                                                            mention information they later provided after question-
             We frequently found that respondents associated patrols                               naires were completed (e.g. borrowing guns). Overall, it is
             with corruption. One individual stated ‘patrols only come                             likely our findings underestimate hunting prevalence.
             to catch the money, not to stop people’, and another stated                               Although specialized methods such as the unmatched
             ‘patrols only use laws for villagers, they have different rules                       count technique can be useful for reducing biases asso-
             for outsiders or people with power’. One respondent, who                              ciated with sensitive topics, they require careful design,
             was a commercial hunter, reported they avoided punish-                                extensive piloting and are not suitable for all contexts
             ment because they were on friendly terms with rangers,                                (Hinsley et al., ). Greater understanding of the relia-
             and other respondents reported that if hunters were caught,                           bility of these methods for providing robust estimates of
             rangers would ask for or accept a bribe. Others believed that                         rule-breaking behaviour is needed. Typically, estimates
             when rangers confiscated meat, they ate it themselves in-                             derived from specialized questioning techniques are vali-
             stead of destroying it. After one interview a respondent                              dated by comparing them to those obtained from direct
             reported they had found a muntjac fawn in the forest, but                             questioning; if prevalence estimates from the specialized
             when urged to take it to the nearest patrol station, they                             method are higher, the method is perceived as more success-
             refused, as they believed that the rangers may eat it. No-                            ful. In addition to undermining the anonymity of the meth-
             body explicitly reported having paid a bribe when caught                              od (Ibbett & Brittain, ), this validation approach fails
             hunting, but two respondents said they had paid bribes to                             to inform researchers whether respondents understood the
             patrols when transporting wood, and others reported they                              method and felt sufficiently protected to report their behav-
             had heard that other villagers paid bribes to release confis-                         iour accurately. Typically, conservation research focuses on
             cated motorbikes. One respondent said that bribery occurs                             obtaining data to answer urgent questions, rather than test-
             because the low-paid rangers have to pay their superiors, to                          ing methods per se. Yet, experimental studies that explictly
             maintain their positions.                                                             assess methods such as the unmatched count technique
                                                                                                   would enhance research practice and improve the reliability
                                                                                                   of data used to inform conservation interventions.
             Discussion                                                                                Snaring has been identified as a specific threat to
                                                                                                   Cambodia’s wildlife (Harrison et al., ; O’Kelly et al.,
             Hunting is widely cited as a cause of the loss of biodiver-                           ) but, although snares were widely used to protect
             sity in Cambodia (Harrison et al., ; Gray et al., ).                          crops, few respondents reported setting snares to hunt wild-
             Our research confirms that local communities living in                                life in the forest. Snaring by ex-hunters, who theoretically
             Keo Seima Wildlife Sanctuary do hunt, although there is                               have less incentive to misreport behaviour, was also low.

                                                                                                Oryx, Page 8 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
Estimating hunting in Cambodia                     9

           We suspect that only a few individuals in each village hunt                           contributed to increasing debt burdens within local com-
           for commercial purposes, probably facilitated by middle-                              munities (Mahanty & Milne, ). Our findings suggest con-
           men who may place orders, purchase catch, and in some                                 sumption of wild meat is widespread but low, comprising
           cases supply equipment, as documented elsewhere in                                    only a few meals per month. Historically, forest products
           Cambodia (Gray et al., ; Coad et al., ). One limita-                          such as wildlife have provided communities with a safety
           tion of our research is that our study was restricted to the                          net in times of hardship (Milner-Gulland et al., ) but
           hunting activity of local people, although hunting is also                            economic vulnerability associated with growing debt may
           thought to be undertaken by Vietnamese nationals near                                 result in increased pressure on natural resources, includ-
           the international border (O’Kelly et al., ), by logging                           ing wildlife. This could be exacerbated by infrastructural
           gangs who stay in the forest for extended periods (HI,                                improvements, which may enhance local, provincial and
           pers. obs., ), and by military or police personnel with                           regional access to wildlife trade. Intelligence-gathering opera-
           high-powered rifles (Drury, ; Evans et al., ). Gath-                          tions that assess commodity chains would be beneficial for
           ering information on prevalence of hunting by these groups                            an understanding of demand dynamics, but in-depth under-
           should be a research priority, although such research could                           standing of the norms and attitudes driving consumption is
           potentially pose significant risk to researchers.                                     also required. To be effective, any intervention for behaviour-
               Our findings highlight a threat to wildlife from hunting                          al change must be informed by robust evidence, and include
           by domestic dogs, an issue identified elsewhere in Cambodia                           appropriate monitoring and impact evaluation (Veríssimo &
           (Heng et al., ; Loveridge et al., ). The scattered distri-                    Wan, ).
           bution of villages, combined with the frequent accompani-                                 Central to the success of protected areas is that rules gov-
           ment of hunters by dogs, means that interactions between                              erning natural resource use are widely known (Keane et al.,
           dogs and wildlife are likely to occur regularly throughout                            ) and, once known, adhered to (Arias, ). Patrols are de-
           the protected area. The presence of dogs may also have indir-                         ployed to catch those who commit offences and to act as a de-
           ect impacts on wildlife, by inducing fear in wild animals, in-                        terrent to potential offenders (Dobson et al., ). Regardless
           creasing competition for resources with wildlife, and trans-                          of whether law enforcement is effective, if the perceived likeli-
           mitting disease (Gompper, ). To understand the potential                          hood and cost of being caught are high, we would expect peo-
           threats that dogs pose to wildlife more information is needed                         ple should be less likely to offend. Our findings suggest the
           on the ranging and hunting behaviour of dogs (e.g. by using                           effectiveness of patrols as a strategy to reduce hunting varied.
           faecal analysis or GPS tracking), alongside socially acceptable                       The perceived likelihood of being caught was low, but the per-
           interventions that promote responsible dog ownership. For                             ceived likelihood of incurring a punishment if caught was
           example, preliminary surveys by the Wildlife Conservation                             high. These factors combined were sufficient to deter some in-
           Society suggest communities are concerned about the high                              dividuals from hunting, and caused others to develop patrol-
           dog population but lack the means to deliver humane ster-                             avoidance strategies. Yet, rangers were also perceived to un-
           ilization. A free dog sterilization programme could be one                            justly punish local people, although some considered rangers’
           possible solution, although a high proportion of dogs would                           malleability, in particular their alleged willingness to accept a
           need to be sterilized to achieve a sufficient reduction in the                        bribe, to be advantageous. Although we agree with the rec-
           population (O. Griffin, unpubl. data).                                                ommendations of others (more efficient and intelligence-led
               Traditionally, hunting in Keo Seima Wildlife Sanctuary                            patrolling is needed, legislation that criminalizes hunting
           was conducted during resin collection (Drury, ). How-                             and possession of technologies such as snares is required,
           ever, since  resin collection has declined markedly                               and all aspects of judiciary systems need to be strengthened;
           (Cheetham, ), partly because of reduced profitability                             Gray et al., ), we believe that conservation success is un-
           but also because of the loss of resin trees to illegal logging                        likely to be achieved by strengthened law enforcement alone
           and industrial-scale land clearance, which intensified in                             (Travers et al., ). Any approach must be informed by ad-
           c. . Cash cropping has emerged as the primary form                                equate understanding of the drivers of non-compliant be-
           of income generation (Travers et al., ; Mahanty &                                 haviour, together with clear recognition of the incentives
           Milne, ), and therefore many have less time to develop                            most likely to encourage behavioural change.
           hunting skills and spend time hunting, and more income to
           buy wild meat (Coad et al., ). Looking ahead, infrastruc-                         Acknowledgements We thank Nao Sok, Sovanndara Than and
                                                                                                 Sovann Nharak for collecting data; Wildlife Conservation Society
           tural improvements, such as paved roads and improved                                  Cambodia and the Royal Government of Cambodia’s Ministry of
           cellular networks, may further enhance the market integra-                            Environment for their support, facilitation and input into the research.
           tion of villages within the Sanctuary (Riggs et al., ),                           Research was funded by the UK Natural Environment Research
           boosting the prevalence of cash cropping and reducing eco-                            Council (Grant No. NE/N001370/1).
           nomic reliance on traditional livelihood activities.                                  Author contributions Research conception and design: all authors;
               Fluctuations in cash crop prices, declining soil fertility,                       data collection and analysis, lead writing: HI; writing and revision: all
           and high costs of fertilizers, pesticides and land rents, has                         authors.

           Oryx, Page 9 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
10         H. Ibbett et al.

             Conflicts of interest Keo Seima Wildlife Sanctuary receives finan-                                     F O R E S T R Y A D M I N I S T R AT I O N () Chapter —Conservation of
             cial and technical support from the Wildlife Conservation Society                                          Wildlife. Law on Forestry. Articles –. Royal Government of
             Cambodia Programme, with whom OG is affiliated.                                                            Cambodia, Phnom Penh, Cambodia.
                                                                                                                    G LY N N , A.N. () What can we learn with statistical truth serum?
             Ethical standards Research permission was granted by the Royal                                             Design and analysis of the list experiment. Public Opinion Quarterly,
             Government of Cambodia, ethics clearance approved by the Uni-                                              , –.
             versity of Oxford’s Social Sciences & Humanities Interdivisional                                       G O M P P E R , M.E. () The dog–human–wildlife interface: assessing
             Research Ethics Committee (Ref No: R43030/RE004), and the research                                         the scope of the problem. In Free-ranging Dogs and Wildlife
             otherwise abided by the Oryx guidelines on ethical standards.                                              Conservation (ed. M.E. Gompper), pp. –. Oxford University
                                                                                                                        Press, Oxford, UK.
             Data availability Data are available from reshare.ukdataservice.ac.                                    G R A Y , T.N.E., S O K U N , H., L E F T E R , E., G R O S U , R., K I M S R E N G , K.,
             uk/854333.                                                                                                 O M A L I S S , K. & G A U N T L E T T , S. () A decade of zero elephant
                                                                                                                        poaching in the Cardamom Rainforest Landscape, Cambodia.
                                                                                                                        Gajah, , –.
             References                                                                                             G R A Y , T.N.E., H U G H E S , A.C., L A U R A N C E , W.F., L O N G , B., L Y N A M , A.
                                                                                                                        J., K E L LY , H.O. et al. () The wildlife snaring crisis: an insidious
             A R I A S , A. () Understanding and managing compliance in the                                         and pervasive threat to biodiversity in Southeast Asia. Biodiversity
                nature conservation context. Journal of Environmental                                                   and Conservation, , –.
                Management, , –.                                                                           G R I F F I N , O. () The Biodiversity of Keo Seima Wildlife Sanctuary
             B E N I T E Z -L O P E Z , A., A L K E M A D E , R., S C H I P P E R , A.M., I N G R A M , D.J.,           ចម្រុះនៅក្នុងដែនជ ម្រកសត្វព្រៃកែវសីមា. Technical Report.
                V E R W E I J , P.A., E I K E L B O O M , J.A.J. et al. () The impact of                            Wildlife Conservation Society, Phnom Penh, Cambodia. dx.doi.org/
                hunting on tropical mammal and bird populations. Science,                                               ./RG.... [accessed  March ].
                , –.                                                                                       H A R R I S O N , R.D. () Emptying the forest: hunting and the
             B L A I R , G., I M A I , K., P A R K , B., C O P P O C K , A. & C H O U , W. ()                       extirpation of wildlife from tropical nature reserves. Bioscience,
                List: Statistical Methods for the Item Count Technique and List                                         , –.
                Experiment. R Foundation for Statistical Computing, Vienna,                                         H A R R I S O N , R.D., S R E E K A R , R., B R O D I E , J.F., B R O O K , S., L U S K I N , M.,
                Austria. rdrr.io/cran/list [accessed  March ].                                                    K E L LY , H.O. et al. () Impacts of hunting on tropical forests
             B O R G E R S O N , C. () The effects of illegal hunting and habitat on two                            in Southeast Asia. Conservation Biology, , –.
                sympatric endangered primates. International Journal of                                             H E N G , S., D O N G , T., H O N , N. & O L S S O N , A. () The hairy-nosed
                Primatology, , –.                                                                                 otter Lutra sumatrana in Cambodia: distribution and notes on
             C H E E T H A M , S. () Tapping for Change: Investigating the Dynamics                                 ecology and conservation. Cambodian Journal of Natural History,
                of Resin Tapping in Seima Protection Forest, Cambodia. Imperial                                         , –.
                College London, London, UK.                                                                         H I N S L E Y , A., N U N O , A., K E A N E , A., S T J O H N , F.A.V. & I B B E T T , H.
             C OA D , L., L I M , S. & N U O N , L. () Wildlife and livelihoods in the                              () Asking sensitive questions using the unmatched count
                Cardamom Mountains. Frontiers in Ecology and Evolution, , .                                         technique: applications and guidelines for conservation. Methods
             D A LT R Y , J. () Editorial—Cambodia’s biodiversity revealed.                                         in Ecology and Evolution, , –.
                Cambodian Journal of Natural History, , –.                                                    H U G H E S , A.C. () Understanding the drivers of Southeast Asian
             D O B S O N , A.D.M., M I L N E R -G U L L A N D , E.J., B E A L E , C.M., I B B E T T , H.                biodiversity loss. Ecosphere, , e.
                & K E A N E , A. () Detecting deterrence from patrol data.                                      I B B E T T , H. & B R I T T A I N , S. () Conservation publications and their
                Conservation Biology, , –.                                                                      provisions to protect research participants. Conservation Biology,
             D R O I T C O U R , J., C A S P A R , R.A., H U B B A R D , M.L., P A R S L E Y , T.L.,                    , –.
                V I S S C H E R , W. & E Z Z AT I , T.M. () The item count technique                            I B B E T T , H., M I L N E R -G U L L A N D , E.J., B E A L E , C., D O B S O N , A.D.,
                as a method of indirect questioning: a review of its development                                        G R I F F I N , O., O’K E L LY , H. & K E A N E , A. () Experimentally
                and a case study application. In Measurement Errors in Surveys                                          assessing the effect of search effort on snare detectability. Biological
                (eds P.P. Biemer, R.M. Groves, L.E. Lyberg, N.A. Mathiowetz &                                           Conservation, , .
                S. Sudman), pp. –. John Wiley & Sons, New Jersey, USA.                                        I N G R A M , D.J., C O A D , L., A B E R N E T H Y , K., M A I S E L S , F., S T O K E S , E.J.,
             D R U R Y , R. () Wildlife Use and Trade in Seima Biodiversity                                         B O B O , K.S. et al. () Assessing Africa-wide pangolin exploitation
                Conservation Area, Mondulkiri Province, Cambodia. University                                            by scaling local data. Conservation Letters, , e.
                College London, London, UK.                                                                         J O N E S , S., K E A N E , A., S T J O H N , F.A.V., V I C K E R Y , J. & P A P W O R T H , S.
             E VA N S , T.D., H O U T , P., P H E T , P. & H A N G , M. () A Study of                               () Audience segmentation to improve targeting of conservation
                Resin-Tapping and Livelihoods in Southern Mondulkiri, Cambodia,                                         interventions for hunters. Conservation Biology, , –.
                with Implications for Conservation and Forest Management. Wildlife                                  K E A N E , A., J O N E S , J.P.G. & M I L N E R -G U L L A N D , E.J. () Encounter
                Conservation Society Cambodia Program, Phnom Penh, Cambodia.                                            data in resource management and ecology: pitfalls and possibilities.
             E VA N S , T.D., A R P E L S , M. & C L E M E N T S , T. () Pilot REDD activities                      Journal of Applied Ecology, , –.
                in Cambodia are expected to improve access to forest resource use                                   K R U M P A L , I. () Determinants of social desirability bias in sensitive
                rights and land tenure for local communities. In Lessons About Land                                     surveys: a literature review. Quality and Quantity, , –.
                Tenure, Forest Governance and REDD+. Case Studies From Africa,                                      L O U C K S , C., M A S C I A , M.B., M A X W E L L , A., H U Y , K., D U O N G , K.,
                Asia and Latin America (eds L. Naughton-Treves & C. Day),                                               C H E A , N. et al. () Wildlife decline in Cambodia, –:
                pp. –. UW-Madison Land Tenure Center, Madison, USA.                                                 exploring the legacy of armed conflict. Conservation Letters,
             E VA N S , T.D., O’K E L LY , H.J., S O R I Y U N , M., H O R , N.M., P H A K T R A , P.,                  , –.
                P H E A K D E Y , S. & P O L L A R D , E. () Seima protection forest. In                        L O V E R I D G E , R., C U S AC K , J., E A M E S , J., S A M N A N G , E. & W I L L C O X , D.
                Evidence-Based Conservation: Lessons From the Lower Mekong                                              () Mammal records and conservation threats in Siem Pang
                (eds T.C.H. Sunderland, J.A. Sayer & H. Minh-Ha), pp. –.                                          wildlife sanctuary and Siem Pang Khang Lech wildlife sanctuary,
                CIFOR, Bogor, Indonesia.                                                                                Cambodia. Cambodian Journal of Natural History, , –.

                                                                                                                Oryx, Page 10 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
Estimating hunting in Cambodia                    11

           M A H A N T Y , S. & M I L N E , S. () Anatomy of a boom: cassava as a                            collapse of the Indochinese leopard (Panthera pardus delacouri)
               ‘gateway’ crop in Cambodia’s north eastern borderland. Asia Pacific                               in Southeast Asia. Biological Conservation, , –.
               Viewpoint, , –.                                                                        S O D H I , N.S., K O H , L.P., B R O O K , B.W. & N G , P.K.L. () Southeast
           M A R T I N , E.B. & P H I P P S , M. () A review of the wild animal trade in                     Asian biodiversity: an impending disaster. Trends in Ecology and
               Cambodia. TRAFFIC Bulletin, , –.                                                            Evolution, , –.
           M I L N E R -G U L L A N D , E.J., B E N N E T T , E.L., A B E R N E T H Y , K., B A K A R R ,     S O LO M O N , J., J A C O B S O N , S.K., W A L D , K.D. & G AV I N , M. ()
               M., B E N N E T T , E., B O D M E R , R. et al. () Wild meat: the bigger                      Estimating illegal resource use at a Ugandan Park with the
               picture. Trends in Ecology and Evolution, , –.                                            randomized response technique. Human Dimensions of Wildlife,
           M O E (M I N I S T R Y O F E N V I R O N M E N T ) () Chapter . Article .                      , –.
               Protected Areas Law. Phnom Penh, Cambodia.                                                     S O U T E R , N.J., S I M P S O N , V., M O U L D , A., E A M E S , J.C., G R AY , T.N.E.,
           M O E (M I N I S T R Y O F E N V I R O N M E N T ) () Sub-Decree No. on                         S I N C L A I R , R. et al. () Editorial: will the recent changes in
               Establishment of Keo Seima Wildlife Sanctuary. Phnom Penh,                                        protected area management and the creation of five new protected
               Cambodia.                                                                                         areas improve biodiversity conservation in Cambodia? Cambodian
           N I E L S E N , M.R., M E I L B Y , H., S M I T H -H A L L , C., P O U L I OT , M. & T R E U E ,      Journal of Natural History, , –.
               T. () The importance of wild meat in the global south. Ecological                          S T J O H N , F.A.V., K E A N E , A.M. & M I L N E R -G U L L A N D , E.J. ()
               Economics, , –.                                                                          Effective conservation depends upon understanding human
           N O S S , A.J. () The impacts of cable snare hunting on wildlife                                  behaviour. In Key Topics in Conservation Biology  (eds D.W.
               populations in the forests of the Central African Republic.                                       Macdonald & K.J. Willis), pp. –. st edition. John Wiley &
               Conservation Biology, , –.                                                                Sons, Oxford, UK.
           N U N O , A. & S T J O H N , F.A.V. () How to ask sensitive questions                          S TA R R , C., N E K A R I S , K.A.I.A.I., S T R E I C H E R , U. & L E U N G , L.K.-P.P.
               in conservation: a review of specialized questioning techniques.                                  () Field surveys of the Vulnerable pygmy slow loris Nycticebus
               Biological Conservation, , –.                                                               pygmaeus using local knowledge in Mondulkiri Province,
           O’KELLY, H.J., E VA N S , T.D., S T O K E S , E.J., C L E M E N T S , T., D A R A , A.,               Cambodia. Oryx, , –.
               G AT E LY , M. et al. () Identifying conservation successes,                               T O U R A N G E A U , R. & Y A N , T. () Sensitive questions in surveys.
               failures and future opportunities; assessing recovery potential of                                Psychological Bulletin, , –.
               wild ungulates and tigers in Eastern Cambodia. PLOS ONE,                                       T R AV E R S , H. () Making friends with benefits: an investigation
               , e.                                                                                         into the use of incentives for the conservation of forest commons.
           O’K E L LY , H.J., R O W C L I F F E , M.J., D U R A N T , S.M. & M I L N E R -                       PhD thesis, Imperial College London, London, UK.
               G U L L A N D , E.J. () Robust estimation of snare prevalence within                       T R AV E R S , H., W I N N E Y , K., C L E M E N T S , T., E VA N S , T. &
               a tropical forest context using N-mixture models. Biological                                      M I L N E R -G U L L A N D , E.J. () A tale of two villages:
               Conservation, , –.                                                                         an investigation of conservation-driven land tenure reform
           R C O R E T E A M () R: a Language and Environment for Statistical                                in a Cambodian Protection Forest. Land Use Policy, ,
               Computing. R Foundation for Statistical Computing, Vienna,                                        –.
               Austria. r-project.org [accessed  March ].                                               T R AV E R S , H., C L E M E N T S , T. & M I L N E R - G U L L A N D , E.J. ()
           R I G G S , R.A., L A N G S T O N , J.D. & S AY E R , J. () Incorporating                         Predicting responses to conservation interventions through scenarios:
               governance into forest transition frameworks to understand and                                    a Cambodian case study. Biological Conservation, , –.
               influence Cambodia’s forest landscapes. Forest Policy and                                      V E R Í S S I M O , D. & W A N , A.K.Y. () Characterizing efforts to reduce
               Economics, , –.                                                                             consumer demand for wildlife products. Conservation Biology,
           RIPPLE, W.J., A B E R N E T H Y , K., B E T T S , M.G., C H A P R O N , G., D I R Z O , R.,           , –.
               G A L E T T I , M. et al. () Bushmeat hunting and extinction risk to                       Y E R S I N , H., A S I I M W E , C., V O O R D O U W , M.J. & Z U B E R B Ü H L E R , K.
               the world’s mammals. Royal Society Open Science, , .                                       () Impact of snare injuries on parasite prevalence in wild
           R O S T R O -G A R C Í A , S., K A M L E R , J.F., A S H , E., C L E M E N T S , G.R.,                chimpanzees (Pan troglodytes). International Journal of
               G I B S O N , L., L Y N A M , A.J. et al. () Endangered leopards: range                       Primatology, , –.

           Oryx, Page 11 of 11 © Fauna & Flora International 2020 doi:10.1017/S0030605319001455
Downloaded from https://www.cambridge.org/core. IP address: 46.4.80.155, on 19 Dec 2020 at 07:37:34, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms
. https://doi.org/10.1017/S0030605319001455
You can also read