Reliability of Non-Expert Observer Estimates of the Magnitude of Marine Reserve Effects

N.S. Hassell, D.H. Williamson, Richard Evans, G.R. Russ

    Research output: Contribution to journalArticlepeer-review

    4 Citations (Scopus)


    No-take marine reserves (NTMRs) are being implemented for various conservation and fishery management objectives. Well-designed ecological monitoring programs should accompany the implementation of NTMRs to assess their performance against stated objectives. However, monitoring programs often have limited funding and limited access to experienced ("expert") underwater visual census (UVC) observers. Consequently, use of non-expert, volunteer observers to collect marine ecological data has increased. The need to assess the quality and utility of non-expert data has been acknowledged. Few studies have directly compared expert and non-expert assessments, and these are generally one-point-in-time comparisons. Here we assess the magnitude of NTMR:Non-reserve density ratios for target and non-target reef fish collected in five separate years over a decade (2000-2009) by expert and non-expert observers in the Palm Islands, Great Barrier Reef, Australia. Despite variations in methods and resultant differences in absolute fish density estimates, NTMR:Non-reserve density ratio estimates made by experts and non-experts usually differed by 10% or less. Furthermore, estimates of the statistical significance of reserve effects made by experts and non-experts had at least 80% agreement over the decade. The findings support the notion that UVC data collected by trained non-expert observers can provide reliable information on the magnitude of NTMR effects. © 2013 Copyright Taylor and Francis Group, LLC.
    Original languageEnglish
    Pages (from-to)361-380
    JournalCoastal Management
    Issue number4
    Publication statusPublished - 2013


    Dive into the research topics of 'Reliability of Non-Expert Observer Estimates of the Magnitude of Marine Reserve Effects'. Together they form a unique fingerprint.

    Cite this