196 R. E. Sykes et al.
coverage and minimizing the number of national and sub- national sampling units. Based on this sensitivity analysis we chose a target value of 10%, as the number of sampling units required tomeet higher targets wasmore than two-fold greater (Supplementary Material 1, Supplementary Table 3). Thus, the set of sampling units identified by Marxan con- tained 10% of the total area of each of the 89 features. The Stage 1 and Stage 2 analyses both involved 1,000
Marxan runs (see Supplementary Material 2 for more de- tails). The Stage 1 analysis was based on 900 national and sub-national sampling units. Each run consisted of 10 mil- lion iterations, and we set the costs so that Marxan ensured each portfolio met all of the targets and also minimized the number of countries selected (Supplementary Material 2). The Stage 2 analysis was based on the 3,377 grid squares found within the national and sub-national sampling units selected in Stage 1. Each run consisted of 100 million itera- tions, and we set the costs so that Marxan ensured each portfolio met all of the targets.
Comparative analyses
To measure whether using our prioritization approach produced better results than sampling units at random, we created 1,000 randomly selected sets of national and sub-national sampling units (analogous to the Stage 1 Marxan analysis) and 1,000 randomly selected sets of the 100 × 100 km sampling units (analogous to the Stage 2 Marxan analysis but based on all of the sampling units across the global terrestrial realm, not only those found within the selected Stage 1 Marxan analysis areas). To do this, we used Python (Van Rossum & Drake, 2009) to ran- domly select sampling units until the set met or exceeded the mean of the combined areas of the 1,000 Stage 1 or Stage 2 Marxan outputs and to calculate the characteristics of the Marxan and random samples. We then undertook three analyses to compare the two
Marxan and two random samples. The first analysis com- pared the extent to which the different samples met the fea- ture targets and therefore represented the different factors linked to drivers of conservation area establishment and biodiversity patterns. The second analysis compared the number of countries and number of Stage 1 (national and sub-national) sampling units selected and therefore the effort needed to collect conservation area data. The third analysis compared the per cent of the terrestrial realm cov- ered by any protected areas. The protected area data came from the publicly available World Database of Protected Areas dataset downloaded in May 2021 (UNEP-WCMC & IUCN, 2021). It should be noted that the publicly available World Database of Protected Areas data does not include most protected areas in China and India. We followed the standard protocol (UNEP-WCMC & IUCN, 2016)by excluding protected areas that are ‘Proposed’ or ‘Not
Reported’ and UNESCO Man and the Biosphere Pro- gramme Reserves. We included point data if the protected area extent was recorded, converting it into a polygon of the required size by producing a buffer with the required radius around the point (UNEP-WCMC & IUCN, 2016). We combined the protected areas for each country, used QGIS to calculate the total area in each grid square and then calculated the overall per cent protected area cover- age for each of the Marxan and random sets.
Results
Stage 1 analysis The best portfolio identified using Marxan comprised nine whole countries and territories and 33 of the sub-national sampling units within another 16 countries (Fig. 2a). These 25 countries and territories are Argentina, Australia, Brazil, China, Democratic Republic of the Congo, Dom- inican Republic, France, French Polynesia, Greenland, India, Indonesia, Italy, Kazakhstan, Kiribati, Mali, Mexico, Papua New Guinea, Russia, Saudi Arabia, South Africa, South Georgia and the South Sandwich Islands, Sudan, Sweden, Tanzania and the USA. We selected only 17 of these 42 sampling units in every one of the 1,000 portfolios identified by Marxan (Fig. 2b), meaning that each of the other 25 sampling units could be swapped for sampling units containing similar amounts of the different features to produce similarly efficient portfolios.
Stage 2 analysis
The best portfolio identified by Marxan met all of the targets and contained 2,231 of the 3,377 sampling units found within the Stage 1 sample, covering 10.9% of the global terrestrial area (Fig. 3a). The combined area of the selected Stage 1 sam- pling units also selected in Stage 2 ranged from 31.5% for Australia to 100% for the Dominican Republic, with a me- dian of 76.1%(Fig. 3a); only seven countries had less than half of their Stage 1 areas selected in Stage 2. The selection frequency results for Stage 2 mirror this pattern, with low scores for sampling units where Marxan only needed to se- lect a smaller proportion of the national and sub-national sampling units (Fig. 3b).
Sampling comparison
The area of the terrestrial realm, excluding Antarctica, in our analysis is 135,008,972 km2. The mean selected area of the 1,000 Stage 1 Marxan outputs was 16.8 ± SD 1.0%of the terrestrial realm and the mean selected area of the 1,000 Stage 2 Marxan outputs was 10.9 ± SD 0.006%. The global area of the different features varied between ,0.001% for the Micronesia sub-region and 94% for
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
