The value of marine ecological data collected by volunteers

  • Published on

  • View

  • Download


  • al

    *, S


    rm 18

    e iscollected by inexperienced people. An assessment was therefore made of the ability of a group of 13 volunteers, recruited by the

    servation of biological diversity inevitably place enor- tant contributions to it, representing a potentially huge

    tionsensus that science has neither the manpower nor thenancial resources to meet the demands that are beingmous burdens on national science budgets (WorldConservation Strategy, 1980). Individual plans areambitious but, while governments, especially those inthe developed world, have allocated some additionalnancial resources to relevant research (e.g. Biodi-versity: The UK Action Plan, 1994), there is widespreadbelief that conservation research is seriously under-fun-ded (e.g. Lewis, 1999). There is almost certainly con-

    workforce. Indeed, they have already made signicantcontributions to scientic knowledge through their par-ticipation in a range of studies, particularly ones thathave been guided by experienced scientists. Examplesinclude surveys of the abundance and distribution ofwaders and wildfowl in British coastal waters (Prater,1981); beached bird surveys (Stowe, 1982); the distribu-tion of Australian birds (Blakers et al., 1984); changes inknowledge and experience. They were capable of performing straight-forward tasks, such as learning to identify species, recordingtheir occurrence on specic parts of the shore and making length measurements of samples of some gastropods. They made some

    recording errors during the eldwork but similar errors were also made by experienced scientists and it is recommended thereforethat all ecological studies should include quality control of data whether or not they involve volunteers. The assessment of abun-dance was problematic. Volunteers assessments for some species were inconsistent and there is evidence that individuals interpreted

    the scale in dierent ways. It is suggested that these problems stemmed from: (1) a lack of eld experience in the volunteers; (2)inadequate guidelines on the use of the abundance scale; and (3) insucient training before eld surveys commenced. However,projects themselves may benet in unexpected ways from the input of volunteers. They contributed taxonomic and computing skillsto the current project. Members of the group also oered new insights by developing hypotheses relating to populations of gas-

    tropods during their eldwork. These were tested and could have formed the basis of longer-term research programmes. There werealso educational benets for the volunteers who were involved in the project. These included increased knowledge of marine issuesand clearer understanding of the ways in which scientic studies are undertaken.

    # 2003 Elsevier Science Ltd. All rights reserved.

    Keywords: Volunteer data; Validation; Quality control; Littoral species mapping; Conservation

    1. Introduction

    National and international programmes for the con-

    However, much of the research that is needed to fullbiodiversity action plans is labour intensive but techni-cally straight-forward. Volunteers could make impor-organisms on shores of the Isle of Cumbrae, Scotland. TheEarthwatch Institute, to collect valid data in a project that aimed to map the distribution and abundance of common littoral

    volunteers ranged considerably in age, educational background,The value of marine ecologic

    Judy Foster-Smith

    The Dove Marine Laboratory (The School of Marine Science and Tech

    Received 28 August 2002; received in revised fo


    Volunteers are potentially a huge scientic resource but ther

    Biological Conservadata collected by volunteers

    tewart M. Evans

    y, Newcastle University), Cullercoats, Tyne and Wear NE30 4PZ, UK

    November 2002; accepted 18 November 2002

    scepticism among some scientists about the reliability of data

    113 (2003) 199213 Australia (Jacoby et al., 1997; Wescott, 1998); NorthSea pollution studies (Evans et al., 2000) and monitor-ing juvenile lobsters (Ellis and Cowan, 2002). The

    0006-3207/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.doi:10.1016/S0006-3207(02)00373-7

    * Corresponding author. Te.l: +44-191-252-4850; fax: +44-191-


    E-mail address: (J. Foster-Smith).placed upon it (e.g. Hodgson, 1999).

    benthic and pelagic communities in Jakarta Bay due toorganic pollution from the city of Jakarta (Harger,1988); coral reef surveys in Singapore (Chou, 1994) andBelize (Mumby et al., 1995); coastal zone management

  • the distributions or abundances of species on the shore.As it turned out, several hypotheses were postulated.

    squares contained areas of both kinds of substrate, biology also varied in the group. One participant had a

    gicalTwo of them, one relating to the sizes of dogwhelksNucella lapillus on parts of the shore that had beenrecently re-colonised following tributyltin (TBT) pollu-tion, and the other to dierences in population structurein the top shells Gibbula cinerea and Gibbula umbilicalis,were tested by making length measurements of samplesof these species.

    2. Methods

    The eld study took place between 9 and 16 July,2001. It was based at the University Marine BiologicalStation, Millport, Scotland.

    2.1. The study site

    The Isle of Cumbrae is located in the Clyde Sea,adjacent to the town of Largs on the west coast ofScotland. The circumference of the island is approxi-mately 18 km, and the town of Millport, which hasabout 2000 inhabitants, is located at the southern end ofthe island. For purposes of the study, the coastal areawas divided into 16 one km recording squares based onthe Ordnance Survey map of the area (Fig. 1). Four tinyareas of shore, between the squares 1 and 16, 6 and 7, 12and 13, and 14 and 15, were excluded from the survey.The Cumbrae shore consists of rocky outcrops that areinterspersed with areas of soft sediment. There were noareas of rock in Square 8, which consisted of part of thesediment shore of Kames Bay. Each of the remainingpotential of this workforce is especially well-illustratedin the coral programme Reef Check (Hodgson, 1999).Recreational divers surveyed over 300 reefs in 31 coun-tries in a global survey that was certainly beyond theresources of conventional scientic projects.Nevertheless, some members of the scientic commu-

    nity are sceptical about the reliability of data generatedby volunteers (Darwell and Dulvy, 1996; Saunders,2002). There is a need therefore to investigate the kindsof tasks that can be performed successfully by inexper-ienced people. The present study addresses this questionin a project in which volunteers collected data to mapthe distribution and abundance of common littoralorganisms on the Isle of Cumbrae, Scotland. In order todo this, they carried out two tasks: (1) the identicationof a range of littoral species; and (2) assessments of theirdistribution and abundance on the shore. However,opportunities were also taken to assess the abilities ofvolunteers to nd organisms on the shore and to learnto identify a range of study species. In addition, thevolunteers were encouraged to formulate hypothesesthat might explain any dierences that they observed in

    200 J. Foster-Smith, S.M. Evans / BioloFig. 1. Map of Cumbrae showing places that are named in the text

    and recording squares, which are 1 km square.

    Conservation 113 (2003) 199213although sediment patches in some of them were verysmall. The recording squares were grouped into fourquartiles (each of four squares), covering the northeast(recording squares 14), southeast (58), southwest (912) and western (1316) parts of the island (see later).The work programme was based primarily on eld-

    work but there were some lectures, covering topics suchas biodiversity, the use of indicator species to assessenvironmental changes caused by natural and anthro-pogenic impacts and (from the skipper of one of theMarine Stations research vessels) an account of the for-mer British basking shark shery. The volunteers weregiven a talk on scientic literature and photocopies ofeight papers relating to the ecology of the study species.

    2.2. The volunteers

    There were 13 volunteers, ve males and eightfemales, recruited by the Earthwatch Institute (see Gil-mour and Saunders, 1995). They contributed nanciallyto the project to cover their travel, accommodation andsubsistence costs. Volunteers ranged in age from 17 to61, and came from a number of dierent backgrounds,including an sixth-form high school student, a recentgraduate, a housewife, a civil engineer, a retired biologyteacher, a personal assistant in a small rm and anemployee of a major oil company. Knowledge of marine

  • (Table 1) at the start of the project, beforeundertaking any eldwork, and at the end of it,

    control procedure (see later). They were selected pri-marily because they are easy to identify in the eld.

    at the

    1. Please assess the extent to which you feel a responsibly for protecting the


    that y

    ely; c


    ely; c


    ee wi


    ee wi


    t be p

    J. Foster-Smith, S.M. Evans / Biological Conservation 113 (2003) 199213 201Very strongly; strongly; to a lim

    2. When you next go to a coastal area, how do you rate the possibility


    Certainly; probably; unlik

    3. Do you think that you are likely to join another ecological research s

    Certainly; probably; unlik

    4. In your opinion which of the following best ts the statement that in

    studies, such as surveys of the seashore? You:

    Strongly agree with it; agree with it; disagr

    5. In your opinion, which of the following best ts the statement that o

    eld work? You:

    Strongly agree with it; agree with it; disagr

    6. How well do you understand the term biodiversity?

    Completely; almost compl

    7. List ve scientic journals in which research on marine ecology mighTable 1

    The questionnaire which the participants were asked to complete bothmapping project (non-target species).(iii) Field exercise. The rst day of the eld pro-when the eldwork had been completed.(ii) Species recognition. The participants knowledge

    of littoral species was tested at the start of theproject before any eldwork was undertaken andat the end of the project. This was done byshowing them 40 projected slides of species andasking them to write down either the common orlatin name of each of them. A single mark wasgiven for a fully correct answer (e.g. eitherbeadlet anemone or Actinia equina) and halfmark for a partially correct response (e.g. ane-mone or Actinia). The selection of species, ofwhich slides were shown, consisted of: (i) thosethat were the focus of the mapping project(referred to as control and target speciesbelow); and (ii) those that were likely to beencountered on the shore but were not part of theThey were: the lugworm Arenicola marina, whose pre-sence on the shore can be recognised by the character-istic faecal casts without digging into the sediment, thecommon mussel Mytilus edulis, the dogwhelk Nucellalapillus and the yellow lichen Xanthoria parietina. Eachis present in large numbers on the shore around mostparts of the island. The remaining 16 species, the targetspecies, were assessed by volunteers only. They were:the beadlet anemone Actinia equina, the knotted wrackAscophyllum nodosum, the common starsh Asteriasrubens, the common whelk Buccinum undatum, theedible crab Cancer pagurus, the green shore crab Carci-nus maenas, the gut weed Enteromorpha intestinalis, theserrated wrack Fucus serratus, the edible periwinkleLittorina littorea, the brittlestar Ophiothrix fragilis, thegrey top shell Gibbula cineraria, the at top shell Gib-bula umbilicalis, the hermit crab Pagurus bernhardus, thebuttersh, Pholis gunnellus, the purple sea urchinPsammechinus miliaris and the tortoiseshell limpet Tec-

    start and at the end of the project

    marine environment:

    extent; hardly at all; not at all

    ou will visit the seashore to investigate the plants and animals living

    ertainly not; dont know

    in the future?

    ertainly not; dont know

    rienced volunteers can make signicant contributions to ecological

    th it; strongly disagree with it; dont know

    tunities should exist for everyone to become be involved in ecological

    th it; strongly disagree with it; dont know

    partly; vaguely; not at all in marine biology and two had degrees in biolo-gical sciences. The remainder ranged from self-claimedtotal ignorance of the subject to a general interest inwildlife, including the marine environment. Never-theless, members of the group functioned as a coherentunit during the project, both at work and socially. Theywere highly committed to the project and were preparedto carry out tasks allocated to them as required. Thiswas in spite of poor weather conditions, which consistedmostly of rain and high winds during the study period.

    2.3. Preliminary tasks

    (i) Questionnaire. The participants were asked torespond independently to a questionnaire

    gramme was devoted primarily to trainingvolunteers both to identify common littoralorganisms and to familiarise themselves withabundance assessments (in preparation for themapping exercise; see later). They were alsoasked to nd as many dierent organisms as theycould during a 1-h search of the shore at low tideat White Bay. Specimens were taken back to thelaboratory for identication as necessary.

    2.4. Mapping the distribution and abundance of species

    Distribution and abundance data were collected on 20species. Four of them, the control species, were assessedby both volunteers and scientists, as part of the quality

  • nonparametric tests (i.e. MannWhitney U test; Wil- scientists (SME).

    gicaltura tessulata. Each of the target species was found onat least some parts of the shore during a preliminarysurvey by two of the scientists (SME and JF-S) in June,2001.For safety reasons, volunteers carried out shore sur-

    veys in small groups. There were ve pairs and onegroup of three people. Each volunteer kept independentrecords of the abundance of the four control species andtwo or three additional target species. The latter werechosen from a selection of target species recommendedby the volunteers themselves. The instructions given tothem were to search as much of the area of eachrecording square as possible within approximately anhour, continually assessing and re-assessing the abun-dances of their allocated species. A common sense,broad brush approach to the abundance scale (Super-abundant; Abundant; Common; Frequent; Occasional;Rare; Not Recorded) was used, based on the level ofsearching eort and the relative likelihood of nding aspecies at a particular site. Volunteers were asked tomake their assessments of the abundances of species,without colluding with other members of the group.Each group made assessments in two 1 km recordingsquares per day, over the period of low water, com-pleting half of the Island (eight squares) in the periodof 4 days that was allocated to the surveys. Group 1volunteers (working in three pairs) were allocated thenortheast (squares 14) and southwest (912) quar-tiles, and Group 2 volunteers (two pairs and a three-some), the sou...


View more >