Biodiversity Monitoring Study

About  Contacts  Shop  Links  Privacy


Biodiversity Monitoring Feasibility Study

A methodology for community-based monitoring is presented in the Vegetation Monitoring Manual, A Step-by-Step Guide to Monitoring Native Vegetation in the ACT by Sarah Sharp and Lori Gould. An ACT Environment Grant supported the Molonglo Catchment Group undertaking a feasibility study by consulting with relevant experts to establish the appropriateness of implementing the methodology on a broad scale within the ACT as another means of assessing catchment health.

Volunteers were trained in the use of the methodology presented in the Monitoring Manual and by making consultants available, extended that training to enable the development of plant lists for several sites, which could potentially be included in the ACT portion of a catchment-wide monitoring project. The consultants further assisted volunteers in identifying plant species on request and provide advice to the project steering committee regarding further refinements to the procedures and use of the data over time.

The feasibility of the methodology will be tested in on-ground trials covering:

  • trial runs of methodology by volunteers on several sites to test its usability;
  • assessing the appropriate frequency and timing of surveys;
  • assessing if the methodology is comprehensive enough to capture different vegetation communities; and
  • analysis of trial run data, including:
    • assessing the quality (repeatability, robustness) of the data by having experts assess the same sites as volunteers; and
    • assessing the usefulness of data collected to other stakeholders.

In order to assess the feasibility of utilising the methods in the Manual for usability, repeatability and robustness, the study took a number of vegetation communities - grassland, grassy box woodland, dry forest and revegetation / regeneration sites - with the property owners assessing their own and each other's sites using the methods described in the ACT Vegetation Monitoring Manual.  Each site was also assessed by an independent expert, Rainer Rehwinkel, and the assessment compared to that of each participant.  Rainer then analysed the results and made recommendations.  His conclusions are presented below, and the full report can be downloaded here (  1,697KB)

Conclusions of the analysis of the methods in the Manual

The trial revealed many instances of agreement in many of the analyses of the datasets from the observers that took part in the trial to evaluate the ACT Vegetation Monitoring Manual. This indicates a degree of robustness and repeatability of the methods in the Manual. On the other hand, analyses of some methods revealed a degree of non-agreement between observers. Part of the non-agreement can be explained by the lack of statistical power; in some cases, there were too few data to analyse effectively to detect trends. In the cases where all or a majority of the participants undertook the field procedures at all or most of the sites, dissimilarity was apparent. It should be noted that assignment of statistical probability to the similarity values was not appropriate for the statistics used.

I consider it unlikely that any of the methods would be used for more than four times over a ten year period by a landholder, community group or professional land manager. This pattern of use has been roughly replicated by the four participants at most sites, so the trial could be considered to be a replicate of the sort of intensity of use by a single landholder or land manager at a site over a ten year period. Having undertaken the field procedures and after careful analyses of the field procedures, I conclude that the dissimilarities or sources of error that I noted between the observers can be explained by a variety of factors, as described in analyses of each method in the sections above. Some of these factors have little to do with the methods in the Manual, per se, as discussed below. Other sources of error have been attributed to minor deficits in the Manual and these have been addressed in the Recommendations or Suggestion Boxes in the preceding sections. The recommendations and suggestions are collated in Appendix 1.

A major source of error, particularly in methods that collected data from a 'vegetation unit', is likely to have been due to the different observers not knowing the precise boundaries of the units, and as a result, not visiting every section, particularly of the larger units. Some attributes that may have been present may have been overlooked by observers in these methods.

Some of the variation amongst observers is undoubtedly due to the inability by some to have identified some of the plant species. This issue was compounded by the fact that the trial was undertaken during winter and early spring, seasons that are sub-optimal for plant identification. Time lapses between the observations by different participants may also have contributed to the variation. This highlights the need for monitoring surveys to be strictly undertaken at the same time of the year if repeat surveys are done, as discussed in the Manual.

In a study comparing the variance between various field monitoring methods, amongst one of the variables tested was the variance between teams undertaking monitoring trials (Deutschman et al., 2008). This study tested the variance in detecting species using various methods (i.e., plots, transects, quadrats), between sites, vegetation communities and the various teams undertaking the trial. Deutschman et al., (2008) found that the variability amongst the teams participating in the trial was small for common and easily identified species; as low as 1%. Variability was as high as 31% for one hard-to-identify species. This study also found high variability amongst the teams in their estimates of species richness. The findings in the trial of the Manual echoes the Deutschman et al. (2008) study; much of the variance in this trial was attributed to the participants' beginners' level or even lack of plant identification skills.

In some of the methods of the Manual, a certain level of technical understanding is required. There was use of subjective terms in at least two methods in the Manual. These were identified as possible sources of error amongst the participants.

Another source of non-agreement in this trial was identified as being due to the placement of the transects in the Herbaceous Species Abundance method. It is likely that by not having placed the transects in exactly the same location, different cover values have be detected. A recommendation to change the wording in the Manual to reflect this result is included in Analysis of Indicator 4; It is suggested that 'fixed' (permanently marked) transects will detect changes from year-to-year and a simple modification in the Manual should facilitate this. In monitoring guidelines developed by the NSW Department of Primary Industries (Auld, 2009), the author states that revisiting fixed points will provide a greater confidence that detected changes have indeed occurred over time, rather than subsequent monitoring detecting nothing more than the natural variation at the site.

Each of the analyses, above, were accompanied by a summary of the utility of the respective method for its use by landholders and community groups. Assessments were also made whether the methods are useful for a range of vegetation communities. All methods, with the exception of the Riparian Condition indicator, could be effectively employed in all terrestrial vegetation types.

The ACT Vegetation Monitoring Manual provides a set of sound methods and the tools necessary to enable land managers and landholders to effectively assess and monitor the vegetation on their land. The Manual will encourage a greater understanding of the functioning of vegetation and promote the identification of plants and animals that land holder have on their land. The Manual will also increase understanding of fauna habitat values and condition variables that contribute to the quality of a site. The Manual is useful for both landholders and community groups, with some methods requiring a higher skill level than others. For example, as mentioned above, the Herbaceous Species Abundance method requires a fairly high level of skill. This should not be seen as a deterrent, as the Manual suggests that users keep descriptive notes of the unidentified species recorded. Returning to this sheet in subsequent years may enable previously unidentified plants to be subsequently identified. Thus demonstrates one of the major benefits of this and other the methods in the Manual its importance as an educational aid.

About the Molonglo Catchment Group    Contacts    Links    Privacy
Send mail to webmaster with questions or comments about this web site.
Copyright 2004-2014 Molonglo Catchment Group Inc.
Last modified: 27/03/2014