Skip to main content
Biology LibreTexts

9.3: Challenges to PVA Implementation

  • Page ID
  • \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)

    Lack of adequate data

    Population biologists often require several years of survey data to distinguish long-term population trends from “model noise”—short-term population fluctuations caused by weather and other unpredictable events (Figure 9.6). For that reason, general guidelines suggest that at a minimum, six (Morris and Doak, 2002) to 10 (McCarthy et al., 2003) years’ worth of population data are required before a PVA is attempted.

    Population biologists often require several years of survey data to distinguish long-term population trends from short-term population fluctuations.

    Figure 9.6 It often takes several years of data to distinguish long-term population trends from the “noise” caused by short-term fluctuations. In this example, it appears as if Kenya’s topi (Damaliscus lunatus jimela, VU) population size is relatively stable, and sometimes even increasing, between 1977 and 1989. However, the 82% decline is unmistakable when long-term trends are considered. After Ogutu et al., 2016, CC BY 4.0.

    In recent years, considerable effort has been invested in collating, summarising, and making available demographic datasets. One example is the Demographic Species Knowledge Index (Conde et al., 2019) meant to summarise demographic data obtained from ex situ conservation facilities (Section 11.5). Nevertheless, most African species continue to lack multi-year datasets, while many threatened species lack reliable survey data altogether. Because the enormous task of filling these data gaps is impractical, there is a need to be strategic as to which populations to consider for PVA purposes. For example, it does not make sense to conduct a PVA on each species in a threatened ecosystem when a few carefully selected indicator species will suffice to monitor ecosystem health (McGeoch et al., 2002). Other priorities for PVA efforts include (1) species harvested by humans, (2) species most sensitive to ecosystem changes, (3) species with the greatest uncertainty regarding viability, and (4) species that are the focus of current management efforts (Wilson et al. 2015).

    But even in the absence of reliable and complete datasets, PVAs can still be useful. For example, sensitivity analysis can inform future data collection efforts, particularly to fill gaps that lead to high levels of uncertainty, or to verify data accuracy for particularly sensitive parameters.

    Data reliability

    While strategically filling data gaps should be a priority, it should not come at the expense of data quality and reliability. Many—perhaps most—population monitoring programs are poorly designed (Buckland and Johnston, 2017), leading to biased data, poor survey precision, and misleading results. Poorly designed surveys not only waste valuable time and resources, but the erroneous results also seriously hamper conservation efforts.

    To overcome these shortcomings, there are five criteria that a well-designed monitoring program should satisfy (Buckland and Johnston, 2017). First, survey sites should represent the region or species of interest. Second, a sufficiently large number of monitoring sites should be chosen. Third, monitoring programs should be set up that every target species—whether common or rare—is adequately counted. Fourth, species selected for monitoring should represent the community of interest, rather than charismatic species that are easily detected. Fifth, multiple surveys need to be conducted over time to detect long-term population trends. Given resource constraints, some compromises in survey design may at times be required. It may also be worth considering the use of citizen scientists and new technologies such as camera traps (Section 9.1.4) to improve data collection efficiency and to provide back-up evidence of reported species for follow-up expert review, if needed.

    Model reliability

    While PVAs can provide reasonably accurate predictions when based on reliable data (Brook et al., 2000; McCarthy et al., 2003), many conservationists continue to be sceptical of PVA results and their ability to predict future population changes over time (Crone et al., 2013). Part of the reason is our inability to accurately account for unanticipated future events, such as unusual weather events or the arrival of a new invasive species. There are also mechanistic challenges to PVA modelling, including their sensitivity to model assumptions and slight changes in model parameters i.e. slight changes in model input generate vastly different results. For this reason, some biologists have started to discourage the use of PVAs in conservation management, especially when faced with inadequate data (Ellner et al., 2002).

    While this scepticism is important and model interrogation should always be welcomed (both aspects usually lead to model improvements), PVA will continue to play a crucial role in conservation in the foreseeable future. It is however important for biologists using PVA to be familiar with the challenges associated with model reliability, as well as the assumptions and limitations of each PVA model. It always helps to begin any PVA model with a clear understanding of the ecology of the target population, the threats it faces, and its demographic characteristics, which in turn enables the modellers to better evaluate model results.

    This page titled 9.3: Challenges to PVA Implementation is shared under a CC BY 4.0 license and was authored, remixed, and/or curated by John W. Wilson & Richard B. Primack (Open Book Publishers) via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.