Estimating conservation area coverage 199
national areas as sampling units. Research is needed to test these assumptions and better assess this trade-off between sample size and sampling effort. The best portfolio identified in Stage 1 comprised nine
whole countries and 33 administrative units in a further 16 countries. The selection frequency scores, which are based on how many times each sampling unit was selected in each of the runs, showed that only 17 of these sampling units were chosen every time (Fig. 2b). The other sampling units are potentially interchangeable, which is important because if obtaining data from a particular country was impossible for logistical or political reasons, these units could be excluded and the analysis run again to find suitable replacements (Ball et al., 2009). However, it is likely that some portions of the largest countries will have to be in- cluded to meet all of the targets. The selection frequency results for the Stage 2 analysis also showed potentially inter- changeable sampling units, mostly within the largest sub- national sampling units selected in Stage 1 containing additional land not needed to meet the targets (Fig. 3). This Stage 2 result also shows the efficiency benefits of using a complementarity-based algorithm to select sample areas (Ball et al., 2009), as Marxan was able to meet the 10% targets for each feature in close to 10% of the sampling region, although features belonging to different factors have different spatial distributions and extents. This involved selecting .10% for some features that are found in many of the sampling units and so are over-represented through meeting targets for other features (Table 1, Supplementary Tables 2 & 4). However, this is not expected to affect esti- mates of conservation area coverage based on the Stage 2 sample because the over-represented features include those with both high and low opportunity costs. We found that the Stage 1 and Stage 2 random sets of
sampling units had near-identical levels of protected area coverage to the global figure. However, none of these ran- dom outputs also met all of the feature targets, so they would be less suitable for assessing to which extent a sample of conservation areas represented biodiversity. The Stage 1 and Stage 2 Marxan outputs met all of the feature targets, indicating they could be used to measure conserva- tion area representativeness, but the mean protected area coverage for the Stage 2 outputs is 15.97% compared to the global figure of 15.25% calculated from the publicly available World Database of Protected Areas data. This overestimate could be a result of our sampling framework, and it was unexpected given that this dataset does not include every protected area from China and India. Thus, more research is needed to understand the reasons under- lying this difference, but its impact could be reduced in future by adjusting conservation area estimates from this sampled approach based on the difference between the global and sample World Database of Protected Areas coverage data.
It could be argued that a better approach to choosing a
sample is to select sampling units at random, avoiding the need to make assumptions about which factors drive conservation area extent and global biodiversity patterns. We investigated this and found that the Stage 1 and Stage 2 random sets of sampling units had near-identical levels of protected area coverage to the global figure but would re- quire collecting data from 2–7 times more countries and across 3–12 times more national and sub-national sampling units than the Marxan outputs. Thus, our data collection framework based on minimizing the number of countries selected and minimizing biases in these countries by setting representation targets is more practical.
Policy implications and wider relevance
Ongoing monitoring of progress towards conservation targets is essential, but the required data are often lacking (Brooks et al., 2015). Resolving this will require more re- sources and capacity building (Stephenson et al., 2017), especially at the level of the nation state where most action is carried out and thus where guidance is most needed (Smith et al., 2009). At the same time, we need timely global estimates of progress to inform international policy. Our proposed solution for conservation area coverage is to identify a representative sample of countries and collect better data just from these, taking advantage of the availabil- ity of accurate information that has not yet been officially approved. Importantly, such a study would not need to report the estimated conservation area coverage for each country, avoiding problems associated with reporting data from unofficial national datasets. In this study we have shown that it is possible to identify
such a representative sample of areas from across the globe within a small enough number of countries to make inten- sive data collection realistic. We have demonstrated a proof of concept and identified a sample of a reasonable size that is also a realistic basis for data collection. Our sampling approach is also likely to be suitable for marine conserva- tion areas, as the existing literature suggests that their dist- ributions are similarly impacted by comparable social and socio-economic factors to non-marine conservation areas (Devillers et al., 2020). The next step is to work with local experts to collect ex-
isting conservation area datasets covering the sample we have identified. This should then be used to develop im- proved global conservation area metrics measuring cover- age, connectivity levels (Saura et al., 2018) and how well these conservation area networks represent biodiversity (Butchart et al., 2015). This will be particularly important for other effective area-based conservation measures, as national- and regional-scale data suggest they enhance protected area network connectivity and cover different biodiversity elements (Dudley et al., 2018). Future work
Oryx, 2024, 58(2), 192–201 © The Author(s), 2023. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605323000625
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140
