1 Version 10 (12-Dec-2020): this manuscript is a non-peer reviewed preprint shared via the EarthArXiv

2 server while being considered for publication in a peer-reviewed academic journal. Please refer to the
3 permanent digital object identifier (https://doi.org/10.1038/s41477-020-00813-w).
4
5 Under the Creative Commons license (CC-By Attribution-NonCommercial-NoDerivatives 4.0
6 International) you are free to share the material as long as the authors are credited, you link to
7 the license, and indicate if any changes have been made. You may not share the work in any way
8 that suggests the licensor endorses you or your use. You cannot change the work in any way or
9 use it commercially.

10 The environmental impacts of palm oil in

11 context
12 Erik Meijaard1,2,3*, Thomas Brooks4,5,6, Kimberly M. Carlson7,8, Eleanor M. Slade9, John Garcia Ulloa10,
13 David L.A. Gaveau11, Janice Ser Huay Lee9, Truly Santika1,2, Diego Juffe-Bignoli2,12, Matthew J.
14 Struebig2, Serge A. Wich13,14, Marc Ancrenaz1,15, Lian Pin Koh16, Nadine Zamira17, Jesse. F. Abrams18,19,
15 Herbert H.T. Prins20, Cyriaque N. Sendashonga21, Daniel Murdiyarso10,22, Paul R. Furumo23, Nicholas
16 Macfarlane4, Rachel Hoffmann24, Marcos Persio25, Adrià Descals26, Zoltan Szantoi27,28, Douglas Sheil29
1
17 Borneo Futures, Bandar Seri Begawan, Brunei Darussalam.
2
18 Durrell Institute of Conservation and Ecology, University of Kent, Canterbury, UK.
3
19 School of Biological Sciences, University of Queensland, St Lucia, Australia.
4
20 Science and Knowledge Unit, IUCN, Gland, Switzerland.
5
21 World Agroforestry Center (ICRAF), University of The Philippines Los Baños, Laguna, Philippines.
6
22 Institute for Marine & Antarctic Studies, University of Tasmania, Hobart, Australia.
7
23 Department of Natural Resources and Environmental Management, the University of Hawai’i at
24 Mānoa, HI, USA.
8
25 Department of Environmental Studies, New York University, New York, NY, USA.
9
26 Asian School of the Environment, Nanyang Technological University of Singapore, Singapore.
10
27 Department of Environmental Systems Science, ETH Zürich, Zürich, Switzerland.
11
28 Center for International Forestry Research, Bogor, Indonesia.
12
29 UN Environment Programme World Conservation Monitoring Centre (UNEP-WCMC), Cambridge,
30 UK.
13
31 School of Biological and Environmental Sciences, Liverpool John Moores University, Liverpool, UK.
14
32 Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The
33 Netherlands.
15
34 Kinabatangan Orang-Utan Conservation Programme, Kota Kinabalu, Sabah, Malaysia.
16
35 Department of Biological Sciences, National University of Singapore, Singapore.

8
 17
36 Rainforest Alliance, Washington, DC, USA.
18
37 Department of Ecological Dynamics, Leibniz Institute for Zoo and Wildlife Research, Berlin,
38 Germany.
19
39 Global Systems Institute and Institute for Data Science and Artificial Intelligence, University of
40 Exeter, Exeter, United Kingdom.
20
41 Animal Sciences Group, Wageningen University, the Netherlands.
21
42 IUCN Policy and Programme Group, IUCN, Gland, Switzerland.
22
43 Department of Geophysics and Meteorology, IPB University, Bogor, Indonesia.
23
44 Earth System Science, Stanford University, CA, USA.
24
45 Department of Veterinary Medicine, University of Cambridge, UK.
25
46 Instituto de Ciências Biológicas, Universidade Federal do Pará, Belém, Brazil.
26
47 Centre de Recerca Ecològica i Aplicacions Forestals, Cerdanyola del Vallès, Barcelona, Spain.
27
48 European Commission, Joint Research Centre, Ispra, Italy.
28
49 Stellenbosch University, Stellenbosch, South Africa.
29
50 Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences,
51 Ås, Norway.
52 *e-mail: [email protected]
53 Abstract
54 Delivering the Sustainable Development Goals (SDGs) requires balancing demands on land between
55 agriculture (SDG 2) and biodiversity (SDG 15). The production of vegetable oils, and in particular
56 palm oil, illustrates these competing demands and trade-offs. Palm oil accounts for ~40% of the
57 current global annual demand for vegetable oil as food, animal feed, and fuel (210 million tons (Mt)),
58 but planted oil palm covers less than 5-5.5% of the total global oil crop area (ca. 425 Mha), due to oil
59 palm’s relatively high yields. Recent oil palm expansion in forested regions of Borneo, Sumatra, and
60 the Malay Peninsula, where >90% of global palm oil is produced, has led to substantial concern
61 around oil palm’s role in deforestation. Oil palm expansion’s direct contribution to regional tropical
62 deforestation varies widely, ranging from 3% in West Africa to 47% in Malaysia. Oil palm is also
63 implicated in peatland draining and burning in Southeast Asia. Documented negative environmental
64 impacts from such expansion include biodiversity declines, greenhouse gas emissions, and air
65 pollution. However, oil palm generally produces more oil per area than other oil crops, is often
66 economically viable in sites unsuitable for most other crops, and generates considerable wealth for
67 at least some actors. Global demand for vegetable oils is projected to increase by 46% by 2050.
68 Meeting this demand through additional expansion of oil palm versus other vegetable oil crops will
69 lead to substantial differential effects on biodiversity, food security, climate change, land
70 degradation, and livelihoods. Our review highlights that, although substantial gaps remain in our
71 understanding of the relationship between the environmental, socio-cultural and economic impacts
72 of oil palm, and the scope, stringency and effectiveness of initiatives to address these, there has
73 been little research into the impacts and trade-offs of other vegetable oil crops. Greater research
74 attention needs to be given to investigating the impacts of palm oil production compared to
75 alternatives for the trade-offs to be assessed at a global scale.

8
 76 Over the past 25 years, global oil crops have expanded rapidly, with major impacts on land use1. The
77 land used for growing oil crops grew from 170 million ha (Mha) in 1961 to 425 Mha in 20172 or ~30%
78 of all cropland world-wide3. Oil palm, soy, and rapeseed together account for >80% of all vegetable
79 oil production with cotton, groundnuts, sunflower, olive, and coconut comprising most of the
80 remainder (Table 1, Figure 1). These crops, including soy (125 Mha planted area2) and maize (197
81 Mha planted area2), are also used as animal feed and other products.
82 Oil palm is the most rapidly expanding oil crop. This palm originates from equatorial Africa where it
83 has been cultivated for millennia, but it is now widely grown in Southeast Asia. Between 2008 and
84 2017, oil palm expanded globally at an average rate of 0.7 Mha per year2, and palm oil is the leading
85 and cheapest edible oil in much of Asia and Africa. While it has been estimated that palm oil is an
86 ingredient in 43% of products found in British supermarkets4, we lack comparable studies for the
87 prevalence of other oils.
88 As a wild plant, the oil palm is a colonising species that establishes in open areas. Cultivated palms
89 are commonly planted as monocultures, although the tree is also used in mixed, small-scale and
90 agroforestry settings. To maximize photosynthetic capacity and fruit yields, oil palm requires a warm
91 and wet climate, high solar radiation, and high humidity. It is thus most productive in the humid
92 tropics, while other oil crops, except coconut, grow primarily in subtropical and temperate regions
93 (Table 1). Moreover, because oil palm tolerates many soils including deep peat and sandy substrates,
94 it is often profitable in locations where few other commodity crops are viable. The highest yields
95 from planted oil palm have been reported in Southeast Asia5. Yields are generally lower in Africa6
96 and the Neotropics5, likely reflecting differences in climatic conditions including humidity and cloud
97 cover6, as well as management, occurrence of pests and diseases, and planting stock7.
98 Palm oil is controversial due to its social and environmental impacts and opportunities. Loss of
99 natural habitats, reduction in woody biomass, and peatland drainage that occur during site
100 preparation are the main direct environmental impacts from oil palm development8. Such
101 conversion typically reduces biodiversity and water quality and increases greenhouse gas emissions,
102 and, when fire is used, smoke and haze5,9. Industrial oil palm expansion by large multi-national and
103 national companies is also often associated with social problems, such as land grabbing and conflicts,
104 labour exploitation, social inequity10 and declines in village-level well-being11. In producer countries,
105 oil palm is a valued crop that brings economic development to regions with few alternative
106 agricultural development options12, and generates substantial average livelihood improvements
107 when smallholder farmers adopt oil palm13. Here we review the current understanding of the
108 environmental impacts from oil palm cultivation and assess what we know about other oil crops in
109 comparison. Our focus is on biodiversity implications and the environmental aspects of
110 sustainability, and we acknowledge the importance of considering these alongside socio-cultural,
111 political, and economic outcomes.
112 DEFORESTATION AND OIL PALM EXPANSION
113 A remote sensing assessment found that oil palm plantations covered at least 19.5 Mha globally in
114 2019 (Figure 2), of which an estimated 67.2% were industrial-scale plantings and the remainder
115 smallholders14. With 17.5 Mha, Southeast Asia has the largest area under production, followed by
116 South and Central America (1.31 Mha), Africa (0.58 Mha) and the Pacific (0.14 Mha). However, the
117 actual area under oil palm production could be 10–20% greater than the area detected from satellite
118 imagery, i.e. 21.5–23.4 Mha, because young plantations (< ca. 3 years), open-canopy plantations, or
119 mixed-species agroforests were omitted14. Estimates suggest that the proportion of oil palm area
120 under smallholder cultivation (typically less than 50 ha of land per family15) varies from 30–60% in
121 parts of Malaysia and Indonesia11 to 94% in Nigeria5.

8
122 The overall contribution of oil palm expansion to deforestation varies widely and depends in part on
123 assessment scope (temporal, spatial) and methods. We reviewed 23 studies that reported land use
124 or land cover change involving oil palm (Table S1 and S2). In Malaysian Borneo, oil palm was an
125 important contributor to overall deforestation16. Here, new plantations accounted for 50% of
126 deforestation from 1972 to 2015 when using a 5-year cut-off to link deforestation and oil palm
127 development17 (Figure 3, Figure S2, Table S3). In contrast, one global sample-based study suggested
128 that between 2000 and 2013, just 0.2% of global deforestation in “Intact Forest Landscapes” was
129 caused by oil palm development18.
130 The degree to which oil palm expansion has replaced forests (defined as naturally regenerating
131 closed canopy forests) varies with context. From 1972 to 2015, around 46% of new plantations
132 expanded into forest, with the remainder replacing croplands, pasturelands, scrublands (including
133 secondary forest regrowth), and other land uses5. Individual studies reported forest clearance
134 ranging from 68% of tracked oil palm expansion in Malaysia and 44% in the Peruvian Amazon, to just
135 5–6% in West Africa, Central America, and South America excluding Peru (Figure 3). In general, oil
136 palm expansion in the Neotropics is characterized by the conversion of previously cleared lands
137 instead of forests19,20, although the extent to which oil palm displaces other land uses into forests
138 remains uncertain. In Indonesia and Malaysian Borneo, industrial plantation expansion and
139 associated deforestation have declined since ca. 201121,22. However, smallholder plantings
140 developed to support demand by industrial palm oil mills may be increasing. To date, only two
141 studies have clearly differentiated between forest clearing by smallholders and industrial plantations
142 (Table S2). In Peru, 30% of smallholder plantings resulted in deforestation23, while in Sumatra,
143 Indonesia 39% of smallholder expansion was into forest24. While we still lack broader understanding
144 of the deforestation impacts of smallholders24, recent studies from Indonesian Borneo show that like
145 industrial actors, smallholders sometimes convert fragile ecosystems such as tropical peatlands into
146 oil palm plantations25. Other oil crops have not yet been mapped globally with similar levels of
147 accuracy, precluding detailed assessments and comparisons.
148 OIL PALM’S DIRECT IMPACTS ON SPECIES
149 The International Union for the Conservation of Nature (IUCN) Red List of Threatened Species26
150 documents 321 species for which oil palm is a reported threat, significantly more than for other oil
151 crops (Figure 4, Table 1). Species threatened by oil palm made up 3.5% of the taxa threatened by
152 annual and perennial non-timber crops (9,088 species) and 1.2% of all globally threatened taxa
153 (27,159 species) in 2019 (Supplementary Materials, Table S4). These species include orangutans
154 Pongo spp., gibbons Hylobates spp. and the tiger Panthera tigris. Species threat lists, however, are
155 incomplete as most plant groups have not been comprehensively assessed, and the focus of threat
156 studies may be biased toward certain oil crops. For example, perennial crops (oil palm, coconut,
157 olive) might be more easily identified as a threat to a species than annual crops, because perennial
158 crops facilitate long-term studies that are more difficult with annual crops that may not be planted
159 every year. Also, the IUCN Red List focuses on threats in the recent past, and is thus biased toward
160 crops with recent rapid expansion. Better information is needed for all oil crops about where they
161 are grown, and how their expansion has affected and could affect natural and semi-natural
162 ecosystems and biodiversity. We note that because coconut is primarily grown in tropical island
163 nations it stands out as a particular threat for rare and endemic species with small ranges27 (Table 1).
164 Oil palm plantations contain lower species diversity and abundance for most taxonomic groups
165 when compared to natural forest28,29. Plant diversity in some plantations is less than 1% of that in
166 natural forests28, but because oil palm is perennial, associated plant diversity may exceed that of
167 annual oil crops (Table 1). One study found 298 plant species in the oil palm undergrowth30, and
168 another found 16 species of fern on oil palm trunks31, while a meta-analysis of plant diversity in a
169 range of annual crops, including oil crops, found between one and 15 associated plant species32.

8
170 Plant diversity in any oil croplands also depends on management choices such as tillage, weeding
171 and the use of herbicides or other chemicals.
172 Recorded mammal diversity in oil palm is 47–90% lower than in natural forest33,34, and strongly
173 depends on the proximity of natural forests. Oil palm plantations generally exclude forest specialist
174 species35,36, which are often those species of greatest conservation importance. For example, forest-
175 dependent gibbons (Hylobatidae) cannot survive in stands of monocultural oil palm, but can make
176 use of interspersed forest fragments within an oil palm matrix28. Some species, although unable to
177 survive solely in oil palm, will utilise plantations. For instance, planted oil palm in Malaysian Borneo
178 supported 22 of the 63 mammal species found in forest habitats33, and 31 of 130 bird species37, most
179 of them relatively common species. Oil palm in Guatemala and Brazil supported 23 and 58 bird
180 species, respectively36,38, while 12 species of snakes were found in a Nigerian oil palm plantation39.
181 Various species will enter plantations to feed on oil palm fruit, including Palm-nut Vultures
182 Gypohierax angolensis43 and Chimpanzee Pan troglodytes40 in Africa and porcupines (Hystricidae),
183 civets (Viverridae), macaques (Cercopithecidae), elephants (Elephantidae) and orangutans in
184 Southeast Asia41. The highest diversity of animal species in oil palm areas, however, is generally
185 found in the wider landscape that includes remnant patches of native vegetation42,43. Factors that
186 are likely to positively influence biodiversity values in both industrial-scale and smallholder
187 plantations include higher landscape heterogeneity, the presence of large forest patches and
188 connectivity among these44, and the plant diversity and structure of undergrowth vegetation. For
189 example, in palm areas where there is systematic cattle grazing, bird and dung beetle abundance
190 and diversity increase45,46.
191 Oil palm cultivation involves the introduction and spread of invasive species including the oil palm
192 itself (noted in Madagascar and Brazil’s Atlantic Forests47), as well as non-native cover crops and
193 nitrogen-fixing plants (e.g., Mucuna bracteata or Calopogonium caeruleum). Similarly, management
194 of oil palm plantations can increase the local abundance of species such as Barn Owls Tyto alba,
195 introduced into plantations to control rodents48. Oil palm plantations also support pests such as the
196 Black Rat Rattus rattus, pigs Sus spp., and beetles such as the Asiatic Rhinoceros Beetle Oryctes
197 rhinoceros and the Red Palm Weevil Rhynchophorus ferrugineus49. Such species can impact palm oil
198 production negatively, for example in reducing oil palm yields through damage to the palm or fruit
199 predation50. They also have a range of local effects, both positive and negative for biodiversity,
200 including animals that prey on them, such as snakes, owls, monkeys and cats51, while the extra food
201 provided by oil palm fruits can increase pig populations resulting in reduced seedling recruitment in
202 forests neighbouring oil palm52.
203 Management within oil palm areas to retain riparian reserves and other set-asides containing
204 natural forest may contribute to pollination and pest control within the plantation, although they
205 may also harbour pests and disease53. Studies to date suggest overall limited, or neutral, effects of
206 such set-asides on pest control services, spill over of pest species, or oil palm yield54. There are also
207 plenty of unknowns, for example, the African beetle Elaiedobius kamerunicus has been introduced
208 as an effective oil palm pollinator and is now widely naturalised in Southeast Asia and America
209 where it also persists in native vegetation and visits the inflorescences of native palms but its
210 impacts, if any, are unexamined (DS pers. obs.). No systematic analysis has been conducted to assess
211 the impact of non-native and invasive species associated with other oil crops.
212 Smallholder plantations tend to be smaller and more heterogeneous than industrial developments,
213 which potentially benefits wildlife, but this remains poorly studied29. A handful of studies indicate
214 that smallholdings support a similar number of, or slightly more, bird and mammal species than
215 industrial plantations, e.g. 55. However, species in smallholder plantations may be more exposed to
216 other pressures, such as hunting, when compared to industrial plantations55.

8
217 OTHER ENVIRONMENTAL IMPACTS
218 Oil palm plantations have a predominantly negative net effect on ecosystem functions when
219 compared to primary, selectively logged or secondary forest9. The clearance of forests and drainage
220 of peatlands for oil palm emits substantial carbon dioxide56. Oil palms can maintain high rates of
221 carbon uptake57 and their oil can potentially be used to substitute fossil fuels, and thus contribute
222 towards sustainable energy (SDG 7) and climate change response (SDG 13). Yet, biofuel from oil
223 palm cannot compensate for the carbon released when forests are cleared and peatlands drained
224 over short or medium time-scales (<100 years)58. Moreover, the carbon opportunity cost of oil palm,
225 which reflects the land’s opportunity to store carbon if it is not used for agriculture, is not very
226 different from annual vegetable oil crops58 (Table 1).
227 Oil palm plantations, and the production of palm oil, can also be sources of methane59 and nitrous
228 oxide60, both potent greenhouse gases that contribute further to climate change, although the
229 former is sometimes used as biogas, reducing net greenhouse gas release61. Other emissions
230 associated with oil palm development include elevated isoprene production by palm trees, which
231 influences atmospheric chemistry, cloud cover and rainfall, although how this affects the
232 environment remains unclear62. In addition, there is some evidence that emissions of other organic
233 compounds, e.g., estragole and toluene63, are also higher in oil palm plantations than in forest, but
234 these emissions appear minor compared to isoprene64.
235 Forest loss and land use conversion to oil palm impact the local and regional climate, although the
236 extent of these impacts remains debated65. For example, increased temperatures and reduced
237 rainfall recorded over Borneo since the mid-1970s are thought to relate to the island’s declining
238 forest cover which is partly due to the expansion of oil palm, with climate changes being greater in
239 areas where forest losses were higher66. Indeed, oil palm plantations tend to be hotter, drier and
240 less shaded than forests due to their less dense canopy, and often have higher evapotranspiration
241 rates than forests67. A drier hotter climate increases the risk of fire and concomitant smoke
242 pollution, especially in peat ecosystems68. In addition to human health consequences (e.g.,
243 respiratory diseases, conjunctivitis), such fires can impact wildlife69 and atmospheric processes. For
244 example, aerosols from fires can scatter solar radiation, disrupt evaporation, and promote drought65.
245 Few of these relationships are well-studied.
246 Conversion of natural forests to oil palm plantations increases run-off and sediment export due to
247 loss or reduction of riparian buffers, reduced ground cover, and dense road networks70. Streams
248 flowing through plantations tend to be warmer, shallower, sandier, more turbid, and to have
249 reduced abundances of aquatic species such as dragonflies (Anisoptera) than streams in forested
250 areas71. Fertilizers, pesticides, and other chemicals used on plantations also impact water quality and
251 aquatic habitats72. The effluent from most modern mills is minimized, but release into local rivers
252 has caused negative impacts to people and to aquatic and marine ecosystems73. Some hydrological
253 impacts may be viewed as positive: for example, construction of flood-control channels and
254 sedimentation ponds for palm oil effluent can benefit some water birds74.
255 Drainage of peatlands and other wetlands to establish oil palm disrupts hydrological cycles,
256 potentially impacting neighbouring forests and other habitats75. The protection and restoration of
257 riparian buffers and reserves within oil palm plantations is therefore key to preserving water quality,
258 with recent research also showing the importance of these landscape features for biodiversity and
259 ecosystem function76. Riparian reserve widths required by law in many tropical countries (20–50 m
260 on each bank) can support substantial levels of biodiversity, maintain hydrological functioning, and
261 improve habitat connectivity and permeability for some species within oil palm76. However, research
262 is urgently needed regarding minimum buffer width and size requirements under different contexts,
263 for different taxa, and for different oil crops.

8
264 THE FUTURE OF OIL PALM
265 Demand for agricultural commodities is growing. Some predict that palm oil production will
266 accelerate across tropical Africa77. However, due to current socio-cultural, technical, political and
267 ecological constraints only around one-tenth of the potential 51 million ha in the five main
268 producing countries in tropical Africa is likely to be profitably developed in the near future7,
269 although this might change as technological, financial and governance conditions improve78. The
270 expansion of oil palm in the Neotropics is also uncertain because of greater challenges the sector
271 faces compared to Southeast Asia, including lower yields, high labour costs, volatile socio-political
272 contexts, and high investment costs5. Although the importance of these factors varies from country
273 to country, in general the expansion of the palm oil industry in the Americas depends heavily on
274 economic incentives and policies, and access to international markets.
275 Meeting the growing demand for palm oil1, while adhering to new zero deforestation policies79, and
276 consumer pressure to be more sustainable, will likely require a combination of approaches, including
277 increasing yields in existing production areas especially those managed by smallholders1, and
278 planting in deforested areas and degraded open ecosystems such as man-made pastures57. These
279 strategies span a land-sparing and land-sharing continuum, with higher-yielding oil palm cultivation
280 sparing land and perhaps reducing overall impacts on biodiversity35, although intermediate
281 strategies on the sparing-sharing continuum may be better at meeting broader societal goals80.
282 Irrespective of the optimal strategy, replanting with high-yielding palms or implementing land
283 sharing agroforestry techniques are challenging for smallholders, who often lack resources and
284 technical knowledge, and may not be able to access improved varieties required to increase yields81.
285 In such situations, provision of technical support from government agencies, non-government
286 organisations or private companies may help smallholders choose intensification over clearing more
287 land to increase palm oil production6.
288 The extent to which biofuel demand by international markets will drive oil palm expansion remains
289 unclear. There is resistance from environmental non-governmental organizations and governments,
290 including the European Union, the second-largest palm oil importer after India5, to the use of palm
291 oil as a biofuel to replace fossil fuels and meet climate change mitigation goals. Such resistance is
292 related to the high CO2–emissions from oil palm-driven deforestation and associated peatland
293 development82. Nonetheless, if oil palm is developed on low carbon stock lands, estimates suggest it
294 may have lower carbon emissions per unit of energy produced than other oil crops like European
295 rapeseed83. Consistent and comparable information on the extent and consequences of other oil
296 crops is urgently required to encourage more efficient land use58.
297 GOVERNANCE OPTIONS
298 Efforts to address the impacts of oil palm cultivation and palm oil trade have been the focus of
299 several initiatives. For example, the two main producer countries have set up the Malaysian
300 Sustainable Palm Oil and Indonesian Sustainable Palm Oil certification schemes, which mandate that
301 oil palm producers comply with a set of practices meant to ensure social and environmentally
302 responsible production. International concerns related to deforestation have been addressed
303 through the High Carbon Stock and High Conservation Value approaches84, which are methodologies
304 that guide identification and protection of lands with relatively intact forest or value for biodiversity,
305 ecosystem services, livelihoods and cultural identity. These frameworks are used by producers to
306 meet the requirements of palm oil sustainability initiatives including certification under the
307 Roundtable on Sustainable Palm Oil (RSPO) standard. This standard was recently expanded to
308 include protection, management, and restoration of riparian areas within certified plantations, a
309 prohibition on new planting on peat, and compliance with the standard is now being used to meet
310 corporate zero-deforestation commitments5. There is evidence for positive impacts of RSPO

8
311 certification achieved through improved management practices, including changes in agrochemical
312 use, improved forest protection, and reduced fires and biodiversity losses, although these effects
313 remain small85,86.
314 Many producers and traders of palm oil have now committed to “zero deforestation”. A 2017 cross-
315 commodity survey87 found that companies in the palm oil sector have the highest proportion of no-
316 deforestation commitments across four commodity supply chains (palm oil, soy, timber and cattle)
317 linked to global deforestation. Although most of these commitments have been made by retailers
318 and manufacturers87, oil palm growers have also made such pledges. In 2018, 41 of the 50 palm oil
319 producers with the largest market capitalization and land areas had committed to address
320 deforestation, with 29 of them pledging to adhere to zero deforestation practices88. These
321 commitments have been identified as a factor in declining expansion of oil palm in Malaysia and
322 Indonesia21,22, although low commodity prices have likely also contributed21. Such private supply
323 chain initiatives like certification and zero-deforestation commitments may be most effective in
324 reducing environmental impacts when leveraged with public and institutional support such as
325 plantation moratoria for certain areas and national low-carbon rural development strategies89, as
326 has been demonstrated, for example, in Brazilian soy production90.
327 LAND USE TRADE-OFFS AMONG VEGETABLE OILS
328 While the environmental impacts of oil palm on natural ecosystems are overwhelmingly negative,
329 such impacts also need to be considered in relation to other land uses, including competing
330 vegetable oil commodities, all of which have their own implications for biodiversity, carbon
331 emissions and other environmental dynamics (Table 1). Global vegetable oil production is expected
332 to expand at around 1.5% per year between 2017 and 202791, while use is projected to expand at
333 1.7% per year globally between 2013 and 2050 from a baseline of 165 million tons (Mt), including for
334 use in food, feed and biofuel1. Unless demand for oil decelerates, this implies an additional
335 production of an average of 3.86 Mt of vegetable oil per year. If this production was delivered by oil
336 palm alone, yielding ca. 4 tons of crude palm oil per ha5,92, 31.3 Mha of additional vegetable oil
337 production land would be needed between 2020 and 2050. If, the addition instead all came from
338 soy, yielding about 0.7 tons of oil per ha1, 179 Mha of extra land, or nearly six times as much, would
339 be required. This simple calculation glosses over nuances of substitutability93 or differential yield
340 increases among crops, but illustrates the magnitude of differences between land needed by oil
341 palm and other oil crops94.
342 Understanding impacts is, however, not just a matter of comparing current and projected
343 distributions and yields of different crops and thus land needs, but also requires clarifying how each
344 hectare of land converted to an oil crop impacts both the environment and people. For example, soy
345 is known to have a large negative impact on biodiversity, with few vertebrates occurring in this
346 annual monoculture crop95, and is responsible for loss of high biodiversity savanna and forest
347 ecosystems in South America96. Thus, sustainable development, including simultaneous delivery of
348 SDGs 2 on agriculture and 15 on biodiversity (alongside contributions to SDG 7 on energy and SDG
349 13 on climate), must consider the wider trade-offs posed by sourcing global vegetable oils97. One key
350 uncertainty is the extent to which demand can be met by increasing yields within established
351 vegetable oil croplands. An additional uncertainty is whether other options, for example microalgal-
352 derived lipids98, may soon offer viable alternatives to meet demand for biofuel.
353 THE WAY FORWARD
354 The expansion of oil palm has had large negative environmental impacts and continues to cause
355 deforestation in some regions. Nevertheless, oil palm contributes to economic development5, has
356 improved welfare for at least some people11, and can be consistent with at least some conservation
357 goals especially when compared to other oil crops78. There remain substantial gaps in our

8
358 understanding of oil palm and the interaction between environmental, socio-cultural and economic
359 impacts of the crop, and the scope, stringency and effectiveness of governance initiatives to address
360 these5. None of these concerns and trade-offs are unique to oil palm: they also apply to other
361 vegetable oil crops27,96_ENREF_30, as well as other agricultural products99. Indeed, all land uses and
362 not just those in the tropics have impacts on their environment12, that can either be prevented or
363 restored100. Pressure on the palm oil industry has, however, apparently resulted in more research on
364 the impacts of palm oil production compared to other oils resulting in an urgent need to better study
365 these alternatives.
366 In a world with finite land and growing demands, we must consider global demands for food, fuel
367 and industrial uses hand-in-hand with environmental conservation objectives. Oil palm’s high yields
368 mean that it requires less land to meet global oil demand than other oil crops. However, minimising
369 overall vegetable oil crop impacts requires evaluation for their past, current and projected
370 distribution and impacts, and review of their yields and global trade and uses. This information is
371 needed to enable better planning and governance of land use for all oil crops, matching risks and
372 opportunities with local conditions and realities, and to optimize the simultaneous delivery of the
373 SDGs.
374 LITERATURE CITED
375 1 Byerlee, D., Falcon, W. P. & Naylor, R. L. The Tropical Oil Crop Revolution: Food, Feed, Fuel,
376 and Forests. (Oxford University Press, 2017).
377 2 FAOSTAT. Food and Agriculture Data. http://www.fao.org/faostat/en/#home. (Food and
378 Agriculture Organization of the United Nations, Rome, Italy, 2019).
379 3 Ramankutty, N. et al. Trends in Global Agricultural Land Use: Implications for Environmental
380 Health and Food Security. Ann. Rev. Plant Biol. 69, 789-815, doi:10.1146/annurev-arplant-
381 042817-040256 (2018).
382 4 Independent. The guilty secrets of palm oil: Are you unwittingly contributing to the
383 devastation of the rain forests?. https://www.independent.co.uk/environment/the-guilty-
384 secrets-of-palm-oil-are-you-unwittingly-contributing-to-the-devastation-of-the-rain-
385 1676218.html. (2009).
386 5 Meijaard, E. et al. Oil Palm and Biodiversity – A Situation Analysis. DOI:
387 10.2305/IUCN.CH.2018.11.en. (IUCN Oil Palm Task Force, 2018).
388 6 Woittiez, L. S., van Wijk, M. T., Slingerland, M., van Noordwijk, M. & Giller, K. E. Yield gaps in
389 oil palm: A quantitative review of contributing factors. Europ. J. Agron. 83, 57-77,
390 doi:10.1016/j.eja.2016.11.002 (2017).
391 7 Feintrenie, L., Gazull, L., Goulaouic, R. & Miaro III, L. Spatialized production models for
392 sustainable palm oil in Central Africa: Choices and potentials. Presented at Scaling Up
393 Responsible Land Governance. Annual World Bank Conference on Land and Poverty,
394 Washington DC, March 14-18, 2016. (2016).
395 8 Sheil, D. et al. The impacts and opportunities of oil palm in Southeast Asia. What do we know
396 and what do we need to know? CIFOR Occ. Paper, no. 51 (2009).
397 9 Dislich, C. et al. A review of the ecosystem functions in oil palm plantations, using forests as a
398 reference system. Biol. Rev. 92, 1539-1569, doi:10.1111/brv.12295 (2017).
399 10 Li, T., M. Evidence-based options for advancing social equity in Indonesian palm oil:
400 Implications for research, policy and advocacy. (Center for International Forestry Research
401 (CIFOR), 2018).
402 11 Santika, T. et al. Does oil palm agriculture help alleviate poverty? A multidimensional
403 counterfactual assessment of oil palm development in Indonesia. World Dev. 120, 105-117,
404 doi:10.1016/j.worlddev.2019.04.012 (2019).

8
405 12 Meijaard, E. & Sheil, D. The Moral Minefield of Ethical Oil Palm and Sustainable Development.
406 Front. Forests Glob. Change 2, doi:10.3389/ffgc.2019.00022 (2019).
407 13 Krishna, V., Euler, M., Siregar, H. & Qaim, M. Differential livelihood impacts of oil palm
408 expansion in Indonesia. Agric. Econ. 48, 639-653, doi:10.1111/agec.12363 (2017).
409 14 Descals, A. et al. High-resolution global map of smallholder and industrial closed-canopy oil
410 palm plantations. Preprint at https://essd.copernicus.org/preprints/essd-2020-159/.
411 doi:10.5194/essd-2020-159 (2020).
412 15 RSPO Smallholders Task Force. Smallholders. Retrieved from
413 https://rspo.org/smallholders#definition. (2012).
414 16 Gaveau, D. L. A. et al. Four decades of forest persistence, loss and logging on Borneo. PLOS
415 ONE 9, e101654, doi:10.1371/journal.pone.0101654 (2014).
416 17 Gaveau, D. L. A. et al. Rapid conversions and avoided deforestation: examining four decades
417 of industrial plantation expansion in Borneo. Sci. Rep. 6, 32017, doi:10.1038/srep32017
418 (2016).
419 18 Potapov, P. et al. The last frontiers of wilderness: Tracking loss of intact forest landscapes from
420 2000 to 2013. Sc. Adv. 3, e1600821, doi:10.1126/sciadv.1600821 (2017).
421 19 Vijay, V., Pimm, S. L., Jenkins, C. N. & Smith, S. J. The Impacts of Oil Palm on Recent
422 Deforestation and Biodiversity Loss. PLOS ONE 11, e0159668,
423 doi:10.1371/journal.pone.0159668 (2016).
424 20 Furumo, P. R. & Aide, T. M. Characterizing commercial oil palm expansion in Latin America:
425 land use change and trade. Env. Res. Lett. 12, 024008, doi:10.1088/1748-9326/aa5892 (2017).
426 21 Gaveau, D. L. A. et al. Rise and fall of forest loss and industrial plantations in Borneo (2000–
427 2017). Cons. Lett. 12, e12622, doi:10.1111/conl.12622 (2019).
428 22 Austin, K. G., Schwantes, A., Gu, Y. & Kasibhatla, P. S. What causes deforestation in Indonesia?
429 Env. Res. Lett. 14, 024007, doi:10.1088/1748-9326/aaf6db (2019).
430 23 Gutiérrez-Vélez, V., H. et al. High-yield oil palm expansion spares land at the expense of
431 forests in the Peruvian Amazon. Env. Res. Lett. 6, 044029, doi:10.1088/1748-
432 9326/6/4/044029 (2011).
433 24 Lee, J. S. H. et al. Environmental Impacts of Large-Scale Oil Palm Enterprises Exceed that of
434 Smallholdings in Indonesia. Cons. Lett. 7, 25-33, doi:10.1111/conl.12039 (2014).
435 25 Schoneveld, G. C., Ekowati, D., Andrianto, A. & van der Haar, S. Modeling peat- and forestland
436 conversion by oil palm smallholders in Indonesian Borneo. Env. Res. Lett. 14, 014006,
437 doi:10.1088/1748-9326/aaf044 (2019).
438 26 IUCN. The IUCN Red List of Threatened Species. Version 2019-2. https://www.iucnredlist.org.
439 (Gland, Switzerland, 2019).
440 27 Meijaard, E., Abrams, J. F., Juffe-Bignoli, D., Voigt, M. & Sheil, D. Coconut oil, conservation and
441 the conscientious consumer. Curr. Biol. 30, R757-R758, doi:10.1016/j.cub.2020.05.059 (2020).
442 28 Foster, W. A. et al. Establishing the evidence base for maintaining biodiversity and ecosystem
443 function in the oil palm landscapes of South East Asia. Phil. Trans Roy. Soc. B: Biol. Sc. 366,
444 3277, doi:10.1098/rstb.2011.0041 (2011).
445 29 Savilaakso, S. et al. Systematic review of effects on biodiversity from oil palm production. Env.
446 Evidence 3, 4, doi:10.1186/2047-2382-3-4 (2014).
447 30 Germer, J. U. Spatial undergrowth species composition in oil palm (Elaeis guineensis Jacq.) in
448 West Sumatra, Kommunikations-, Informations- und Medienzentrum der Universität
449 Hohenheim, (2003).
450 31 Sato, T., Itoh, H., Kudo, G., Kheong, Y. S. & Furukawa, A. Species Composition and Structure of
451 Epiphytic Fern Community on Oil Palm Trunks in Malay Archipelago. Tropics 6, 139-148,
452 doi:10.3759/tropics.6.139 (1996).
453 32 Letourneau, D. K. et al. Does plant diversity benefit agroecosystems? A synthetic review. Ecol.
454 Appl. 21, 9-21, doi:10.1890/09-2026.1 (2011).

8
455 33 Wearn, O. R., Carbone, C., Rowcliffe, J. M., Bernard, H. & Ewers, R. M. Grain-dependent
456 responses of mammalian diversity to land use and the implications for conservation set-aside.
457 Ecol. Appl. 26, 1409-1420, doi:10.1890/15-1363 (2016).
458 34 Pardo, L. E. et al. Land management strategies can increase oil palm plantation use by some
459 terrestrial mammals in Colombia. Sci. Rep. 9, 7812, doi:10.1038/s41598-019-44288-y (2019).
460 35 Phalan, B., Onial, M., Balmford, A. & Green, R. E. Reconciling Food Production and Biodiversity
461 Conservation: Land Sharing and Land Sparing Compared. Science 333, 1289-1291,
462 doi:10.1126/science.1208742 (2011).
463 36 Almeida, S. M. et al. The effects of oil palm plantations on the functional diversity of
464 Amazonian birds. J. Trop. Ecol. 32, 510-525, doi:10.1017/S0266467416000377 (2016).
465 37 Edwards, D. P. et al. Selective-logging and oil palm: multitaxon impacts, biodiversity indicators,
466 and trade-offs for conservation planning. Ecol. Applic. 24, 2029-2049, doi:10.1890/14-0010.1
467 (2014).
468 38 Nájera, A. & Simonetti, J. A. Can oil palm plantations become bird friendly? Agrofor. Syst. 80,
469 203-209, doi:10.1007/s10457-010-9278-y (2010).
470 39 Akani, G. C., Ebere, N., Luiselli, L. & Eniang, E. A. Community structure and ecology of snakes
471 in fields of oil palm trees (Elaeis guineensis) in the Niger Delta, southern Nigeria. Afr. J. Ecol.
472 46, 500-506, doi:10.1111/j.1365-2028.2007.00885.x (2008).
473 40 Humle, T. & Matsuzawa, T. Oil palm use by adjacent communities of chimpanzees at Bossou
474 and Nimba Mountains, West Africa. Int. J. Primatol. 25, 551-581,
475 doi:10.1023/B:IJOP.0000023575.93644.f4 (2004).
476 41 Ancrenaz, M. et al. Of pongo, palms, and perceptions – A multidisciplinary assessment of
477 orangutans in an oil palm context. Oryx 49, 465–472, doi:10.1017/S0030605313001270
478 (2015).
479 42 Mitchell, S. L. et al. Riparian reserves help protect forest bird communities in oil palm
480 dominated landscapes. J. Appl. Ecol. 55, 2744-2755, doi:10.1111/1365-2664.13233 (2018).
481 43 Deere, N. J. et al. Implications of zero-deforestation commitments: Forest quality and hunting
482 pressure limit mammal persistence in fragmented tropical landscapes. Cons. Lett. 13, e12701,
483 doi:10.1111/conl.12701 (2020).
484 44 Knowlton, J. L. et al. Oil palm plantations affect movement behavior of a key member of
485 mixed-species flocks of forest birds in Amazonia, Brazil. Trop. Cons. Sc. 10,
486 1940082917692800, doi:10.1177/1940082917692800 (2017).
487 45 Tohiran, K. A. et al. Targeted cattle grazing as an alternative to herbicides for controlling weeds
488 in bird-friendly oil palm plantations. Agron. Sust. Dev. 37, 62, doi:10.1007/s13593-017-0471-
489 5 (2017).
490 46 Slade, E. M. et al. Can cattle grazing in mature oil palm increase biodiversity and ecosystem
491 service provision? The Planter 90, 655-665 (2014).
492 47 IUCN. Global Invasive Species Database (GISD). Species profile Elaeis guineensis. Available
493 from: http://www.iucngisd.org/gisd/species. [Accessed 27 February 2018]. (2015).
494 48 Wan, H. The introduction of barn owl (Tyto alba) to Sabah for rat control in oil palm
495 plantations. Planter 76, 215-222 (2000).
496 49 Bessou, C. et al. Sustainable Palm Oil Production project synthesis: Understanding and
497 anticipating global challenges. (Center for International Forestry Research (CIFOR), 2017).
498 50 Puan, C. L., Goldizen, A. W., Zakaria, M., Hafidzi, M. N. & Baxter, G. S. Relationships among rat
499 numbers, abundance of oil palm fruit and damage levels to fruit in an oil palm plantation.
500 Intergr. Zool. 6, 130-139, doi:10.1111/j.1749-4877.2010.00231.x (2011).
501 51 Holzner, A. et al. Macaques can contribute to greener practices in oil palm plantations when
502 used as biological pest control. Curr. Biol. 29, R1066-R1067, doi:10.1016/j.cub.2019.09.011
503 (2019).

8
504 52 Luskin, M. S. et al. Cross-boundary subsidy cascades from oil palm degrade distant tropical
505 forests. Nature Comms. 8, 2231, doi:10.1038/s41467-017-01920-7 (2017).
506 53 Mayfield, M. M. The importance of nearby forest to known and potential pollinators of oil
507 palm (Elaeis guineensis Jacq.; Areceaceae) in southern Costa Rica. Econ. Botany 59, 190,
508 doi:10.1663/0013-0001(2005)059[0190:TIONFT]2 (2005).
509 54 Woodham, C. R. et al. Effects of replanting and retention of mature oil palm riparian buffers
510 on ecosystem functioning in oil palm plantations. Front. Forests Glob. Change,
511 doi:10.3389/ffgc.2019.00029 (2019).
512 55 Azhar, B. et al. The influence of agricultural system, stand structural complexity and landscape
513 context on foraging birds in oil palm landscapes. Ibis 155, 297-312, doi:10.1111/ibi.12025
514 (2013).
515 56 Wijedasa, L. S. et al. Denial of long-term issues with agriculture on tropical peatlands will have
516 devastating consequences. Glob. Change Biol. 23, 977-982, doi:10.1111/gcb.13516 (2016).
517 57 Quezada, J. C., Etter, A., Ghazoul, J., Buttler, A. & Guillaume, T. Carbon neutral expansion of
518 oil palm plantations in the Neotropics. Sc. Advan. 5, eaaw4418, doi:10.1126/sciadv.aaw4418
519 (2019).
520 58 Searchinger, T. D., Wirsenius, S., Beringer, T. & Dumas, P. Assessing the efficiency of changes
521 in land use for mitigating climate change. Nature 564, 249-253, doi:10.1038/s41586-018-
522 0757-z (2018).
523 59 Reijnders, L. & Huijbregts, M. A. J. Palm oil and the emission of carbon-based greenhouse
524 gases. J. Cleaner Prod. 16, 477-482, doi:10.1016/j.jclepro.2006.07.054 (2006).
525 60 Murdiyarso, D., Van Noordwijk, M., Wasrin, U. R., Tomich, T. P. & Gillison, A. N. Environmental
526 benefits and sustainable land-use options in the Jambi transect, Sumatra. J. Veget. Sc. 13, 429-
527 438, doi:10.1111/j.1654-1103.2002.tb02067.x (2002).
528 61 Harsono, S. S., Grundmann, P. & Soebronto, S. Anaerobic treatment of palm oil mill effluents:
529 potential contribution to net energy yield and reduction of greenhouse gas emissions from
530 biodiesel production. J. Cleaner Prod. 64, 619-627, doi:10.1016/j.jclepro.2013.07.056 (2014).
531 62 Hewitt, C. N. et al. Nitrogen management is essential to prevent tropical oil palm plantations
532 from causing ground-level ozone pollution. Proc. Natl. Acad. Sc. USA 106, 18447,
533 doi:10.1073/pnas.0907541106 (2009).
534 63 Misztal, P. K. et al. Direct ecosystem fluxes of volatile organic compounds from oil palms in
535 South-East Asia. Atmos. Chem. Phys. 11, 8995-9017, doi:10.5194/acp-11-8995-2011 (2011).
536 64 Guenther, A. et al. The Model of Emissions of Gases and Aerosols from Nature version 2.1
537 (MEGAN2. 1): an extended and updated framework for modeling biogenic emissions. (2012).
538 65 Ellison, D. et al. Trees, forests and water: Cool insights for a hot world. Glob. Env. Change 43,
539 51-61, doi:10.1016/j.gloenvcha.2017.01.002 (2017).
540 66 McAlpine, C. A. et al. Forest loss and Borneo's climate. Env. Res. Lett. 13, 044009,
541 doi:10.1088/1748-9326/aaa4ff (2018).
542 67 Fan, Y. et al. Reconciling canopy interception parameterization and rainfall forcing frequency
543 in the community land model for simulating evapotranspiration of rainforests and oil palm
544 plantations in Indonesia. J. Advan. Model. Earth Syst. 11, 732-751,
545 doi:10.1029/2018MS001490 (2019).
546 68 Crippa, P. et al. Population exposure to hazardous air quality due to the 2015 fires in Equatorial
547 Asia. Sci. Rep. 6, 37074, doi:10.1038/srep37074 (2016).
548 69 Nichol, J. Bioclimatic impacts of the 1994 smoke haze event in Southeast Asia. Atmosph. Env.
549 31, 1209-1219, doi:10.1016/S1352-2310(96)00260-9 (1997).
550 70 Carlson, K. M. et al. Consistent results in stream hydrology across multiple watersheds: A reply
551 to Chew and Goh. J. Geophys. Res. Biogeosci. 120, 812-817, doi:10.1002/2014JG002834
552 (2015).

8
553 71 Luke, S. H. et al. The effects of catchment and riparian forest quality on stream environmental
554 conditions across a tropical rainforest and oil palm landscape in Malaysian Borneo. Ecohydrol.
555 10, e1827, doi:10.1002/eco.1827 (2017).
556 72 Mayer, P. M., Reynolds, S. K., McCutchen, M. D. & Canfield, T. J. Meta-Analysis of Nitrogen
557 Removal in Riparian Buffers. J. Env. Qual. 36, 1172-1180, doi:10.2134/jeq2006.0462 (2007).
558 73 Chellaiah, D. & Yule, C. M. Effect of riparian management on stream morphometry and water
559 quality in oil palm plantations in Borneo. Limnologica 69, 72-80,
560 doi:10.1016/j.limno.2017.11.007 (2018).
561 74 Sulai, P. et al. Effects of water quality in oil palm production landscapes on tropical waterbirds
562 in Peninsular Malaysia. Ecol. Res. 30, 941-949, doi:10.1007/s11284-015-1297-8 (2015).
563 75 Anda, M., Siswanto, A. B. & Subandiono, R. E. Properties of organic and acid sulfate soils and
564 water of a ‘reclaimed’ tidal backswamp in Central Kalimantan, Indonesia. Geoderma 149, 54-
565 65, doi:10.1016/j.geoderma.2008.11.021 (2009).
566 76 Luke, S. H. et al. Riparian buffers in tropical agriculture: Scientific support, effectiveness and
567 directions for policy. J. Appl. Ecol. 56, 85-92, doi:10.1111/1365-2664.13280 (2019).
568 77 Wich, Serge A. et al. Will Oil Palm’s Homecoming Spell Doom for Africa’s Great Apes? Curr.
569 Biol. 24, 1659-1663, doi:10.1016/j.cub.2014.05.077 (2014).
570 78 Sayer, J., Ghazoul, J., Nelson, P. & Boedhihartono, A. K. Oil palm expansion transforms tropical
571 landscapes and livelihoods. Glob. Food Secur. 1, 114–119, doi:10.1016/j.gfs.2012.10.003
572 (2012).
573 79 RSPO. RSPO and HCSA Collaborate to Implement No Deforestation in High Forest Cover
574 Landscapes. https://rspo.org/news-and-events/news/rspo-and-hcsa-collaborate-to-
575 implement-no-deforestation-in-high-forest-cover-landscapes. (2018).
576 80 Law, E. A. et al. Mixed policies give more options in multifunctional tropical forest landscapes.
577 J. Appl. Ecol. 54, 51-60, doi:10.1111/1365-2664.12666 (2017).
578 81 Budiadi et al. Oil palm agroforestry: an alternative to enhance farmers’ livelihood resilience.
579 IOP Conf. Ser.: Earth Env. Sc. 336, 012001, doi:10.1088/1755-1315/336/1/012001 (2019).
580 82 Valin, H. et al. The land use change impact of biofuels consumed in the EU. Quantification of
581 area and greenhouse gas impacts. (ECOFYS Netherlands B.V., Utrecht, the Netherlands, 2015).
582 83 Thamsiriroj, T. & Murphy, J. D. Is it better to import palm oil from Thailand to produce
583 biodiesel in Ireland than to produce biodiesel from indigenous Irish rape seed? Appl. Energy
584 86, 595-604, doi:10.1016/j.apenergy.2008.07.010 (2009).
585 84 Rosoman, G., Sheun, S. S., Opal, C., Anderson, P. & Trapshah, R. The HCS Approach Toolkit.
586 (HCS Approach Steering Group, Singapore, 2017).
587 85 Carlson, K. M. et al. Effect of oil palm sustainability certification on deforestation and fire in
588 Indonesia. Proc. Natl. Acad. Sci. USA 115, 121-126, doi:10.1073/pnas.1704728114 (2018).
589 86 Furumo, P. R., Rueda, X., Rodríguez, J. S. & Parés Ramos, I. K. Field evidence for positive
590 certification outcomes on oil palm smallholder management practices in Colombia. J. Cleaner
591 Prod. 245, 118891, doi:https://doi.org/10.1016/j.jclepro.2019.118891 (2020).
592 87 Donofrio, S., Rothrock, P. & Leonard, J. Tracking Corporate Commitments to Deforestation-
593 free Supply Chains, 2017. (Forest Trends, Washington, DC, 2017).
594 88 SPOTT. Palm oil: ESG policy transparency assessments. https://www.spott.org/palm-oil/.
595 (2018).
596 89 Furumo, P. R. & Lambin, E. F. Scaling up zero-deforestation initiatives through public-private
597 partnerships: A look inside post-conflict Colombia. Glob. Env. Ch. 62, 102055,
598 doi:https://doi.org/10.1016/j.gloenvcha.2020.102055 (2020).
599 90 Gibbs, H. K. et al. Brazil's Soy Moratorium. Science 347, 377, doi:10.1126/science.aaa0181
600 (2015).
601 91 OECD and FAO. OECD‑FAO Agricultural Outlook 2018‑2027. (2017).

8
602 92 Johnston, M., Foley, J. A., Holloway, T., Kucharik, C. & Monfreda, C. Resetting global
603 expectations from agricultural biofuels. Env. Res. Lett. 4, 014004, doi:10.1088/1748-
604 9326/4/1/014004 (2009).
605 93 Parsons, S., Raikova, S. & Chuck, C. J. The viability and desirability of replacing palm oil. Nature
606 Sust. 3, 412-418, doi:10.1038/s41893-020-0487-8 (2020).
607 94 Qaim, M., Sibhatu, K. T., Siregar, H. & Grass, I. Environmental, Economic, and Social
608 Consequences of the Oil Palm Boom. Ann. Rev. Res. Econ. 12, 321-344, doi:10.1146/annurev-
609 resource-110119-024922 (2020).
610 95 VanBeek, K. R., Brawn, J. D. & Ward, M. P. Does no-till soybean farming provide any benefits
611 for birds? Agricult. Ecosyst. Env. 185, 59-64, doi:https://doi.org/10.1016/j.agee.2013.12.007
612 (2014).
613 96 Green, J. M. H. et al. Linking global drivers of agricultural trade to on-the-ground impacts on
614 biodiversity. Proc. Natl. Acad. Sci. USA 116, 23202, doi:10.1073/pnas.1905618116 (2019).
615 97 Strona, G. et al. Small room for compromise between oil palm cultivation and primate
616 conservation in Africa. Proc. Natl. Acad. Sci. USA 115, 8811, doi:10.1073/pnas.1804775115
617 (2018).
618 98 Ajjawi, I. et al. Lipid production in Nannochloropsis gaditana is doubled by decreasing
619 expression of a single transcriptional regulator. Nature Biotech. 35, 647, doi:10.1038/nbt.3865
620 (2017).
621 99 De Beenhouwer, M., Aerts, R. & Honnay, O. A global meta-analysis of the biodiversity and
622 ecosystem service benefits of coffee and cacao agroforestry. Agric. Ecosyst. Env. 175, 1-7,
623 doi:10.1016/j.agee.2013.05.003 (2013).
624 100 Strassburg, B. B. N. et al. Global priority areas for ecosystem restoration. Nature,
625 doi:10.1038/s41586-020-2784-9 (2020).
626 101 Payán, E. & Boron, V. The Future of Wild Mammals in Oil Palm Landscapes in the Neotropics.
627 Front. Forests Glob. Change 2, doi:10.3389/ffgc.2019.00061 (2019).
628 102 Maddox, T., Priatna, D., Gemita, E. & Salampessy, A. The conservation of tigers and other
629 wildlife in oil palm plantations Jambi Province, Sumatra, Indonesia. ZSL Conservation Report
630 No.7. (The Zoological Society of London, London, UK, 2007).
631 103 Ancrenaz, M. et al. Pongo pygmaeus. IUCN Red List Threat. Sp., e.T17975A17966347 (2016).
632 104 Pangau-Adam, M., Mühlenberg, M. & Waltert, M. Rainforest disturbance affects population
633 density of the northern cassowary Casuarius unappendiculatus in Papua, Indonesia. Oryx 49,
634 735-742, doi:10.1017/S0030605313001464 (2014).
635 105 Alamgir, M. et al. Infrastructure expansion challenges sustainable development in Papua New
636 Guinea. PLOS ONE 14, e0219408, doi:10.1371/journal.pone.0219408 (2019).
637 106 Katiyar, R. et al. Microalgae: An emerging source of energy based bio-products and a solution
638 for environmental issues. Renew. Sust. Energy Rev. 72, 1083-1093,
639 doi:10.1016/j.rser.2016.10.028 (2017).
640 107 Nomanbhay, S., Salman, B., Hussain, R. & Ong, M. Y. Microwave pyrolysis of lignocellulosic
641 biomass––a contribution to power Africa. Energy Sust. Soc. 7, 23, doi:10.1186/s13705-017-
642 0126-z (2017).
643
644 Correspondence and requests for materials should be addressed to [email protected].
645 Acknowledgements The development of this situation analysis was supported by the IUCN project
646 “Global Commons: Solutions for a Crowded Planet”, funded by the Global Environment Facility. DJB
647 received funding from the UK Research and Innovation’s Global Challenges Research Fund under the
648 Trade, Development and the Environment Hub project (project number ES/S008160/1). MP was
649 supported by the CNPq research productivity fellowships (#308403/2017‐7). JGU was funded by
650 SNSF R4D-project Oil Palm Adaptive Landscapes.

8
651
652 Author contributions
653 EM, DS, and TB conceptualized this study and developed the initial manuscript, with KC, JGU, DG,
654 JSHL, DJB, SAW, MA, SW, LPK, JFA, ZS and AD assisting in the acquisition, analysis, and interpretation
655 of the data and further writing. ES, TS, JA, HP, CS, DM, PF, NM, RH, MP, and MS provided substantial
656 input into the text revisions, and NZ, JA, DJB, KC, DG, AD and JFA designed the graphics.
657
658 Competing Interests statement
659 None of co-authors in this study, except DJB, MP, and JGU, received funding for conducting this
660 review, although the information was partly based on a study funded by the Global Environment
661 Facility. EM, TB, DG, MA, SW, LPK, JGU, KC, NM and DS are members of and have received funding
662 from the IUCN Oil Palm Task Force, a group tasked by the IUCN members to investigate the
663 sustainability of palm oil. TB, DJB, MA, CS and NM work for conservation organizations and EM, MA
664 and MP have done work paid by palm oil companies or the Roundtable on Sustainable Palm Oil.

8
665 FIGURE LEGENDS
666
667 Figure 1. Main vegetable oil crops (see Table 1). (a) Harvested area from 1961 to 2017. (b)
668 Vegetable oil production from 1961 to 2014. Data from FAOSTAT2.

669
670
671

8
672 Figure 2. Maps of industrial and smallholder-scale oil palm from analysis of satellite imagery until
673 the second half of 201914, and examples of species it affects negatively: (a) Panthera onca (Near
674 Threatened)101 and Ara macao (Least Concern)36; (b) Pan troglodytes (Endangered)77; (c) Panthera
675 tigris (Endangered)102, Helarctos malayanus (Vulnerable)102, Pongo pygmaeus (Critically
676 Endangered)103, Casuarius unappendiculatus (Least Concern)104, and Dendrolagus goodfellowi
677 (Endangered)105. The maps lack information on plantations < 3 years old and planted oil palm in
678 mixed agroforestry settings, but provide the most up-to-date estimates available. For each region
679 the percentages of intact (green) and non-intact forests (orange) are shown relative to the total
680 extent of forest ecosystems18.

681
682
683

8
684 Figure 3. Oil palm’s estimated role in deforestation aggregated across studies, years, and regions.
685 Panel a depicts the contribution of oil palm to overall deforestation, while b shows the percentage
686 of all oil palm expansion that cleared forest (Supplementary Methods). There were no data for
687 Peru and South and Central America for panel a, and no global data for panel b. Southeast Asia (SE
688 Asia) excludes Indonesia and Malaysia, which are shown separately, while South America excludes
689 Peru. Each filled circle represents one time period from a single study, with individual studies
690 represented by distinct colours. The size of the circle corresponds to the relative number of area-
691 years represented in that time period (larger circles represent a larger study area and longer time
692 period of sampling). Boxplot middle bars correspond to the unweighted median across study-time
693 periods; lower and upper hinges represent the 25th and 75th percentiles of study-time periods; and
694 whiskers extend from the upper (lower) hinge to the largest (smallest) value no further than 1.5
695 times the interquartile range from the hinge (Figure S2, Tables S2 and S3).

696
697
698

8
699 Figure 4 - Species groups with more than 8 threatened species with the terms "palm oil" or "oil
700 palm" in the threats texts of the IUCN Red List of Threatened Species Assessments26. In total 321
701 species assessments had oil palm plantations as one of the reported threats (301 when excluding
702 groups with < 8 threatened species), which constitutes 3.5% of threatened species threatened by
703 annual and perennial non-timber crops (9,088 species) and 1.2% of all globally threatened species
704 (27,159 species) in 2019 (Supplementary Material and Table S4). CR = Critically Endangered; EN =
705 Endangered; VU = Vulnerable.

706
707
708

8
709 Table 1. Overview of the major oil crops, typical production cycle, yields, main production
710 countries, biomes in which impacts primarily occur, carbon emissions, the number of threatened
711 species according to the IUCN Red List of Threatened Species26 for which the specific crop is
712 mentioned as a threat, and the median species richness and median range-size rarity (amphibians,
713 birds and mammals) of species occurring within the footprint of each crop with first and third
714 quartile in brackets (IUCN Red List) (see Supporting Online Methods, Figure S1, Table S4). Carbon
715 emissions include carbon opportunity costs and production emissions58. “n/a” indicates that no
716 data are available.
Oil crop Type of crop Oil yield Main oil Main biome Kg # species Median Median
(t ha-1) production impacted CO2e/MJ threatened Species range-size
106,107 countries 58 by crop26 Richness rarity (ha
(number ha-1
of 10e5)26
species)26
Oil palm Perennial (25 1.9–4.8 Indonesia, Tropical rainforest 1.2 321 472 [443, 36 [27,
Elaeis years cycle) Malaysia, 504] 57]
guineensis Thailand
Soybean Annual (~6 0.4–0.8 China, USA, Subtropical grass 1.3 73 278 [251, 10 [5, 14]
Glycine max months Brazil, savanna, 462]
cycle), Argentina temperate steppe,
rotated with and broadleaf
other crops forest
Rapeseed Annual (~6 0.7–1.8 China, Temperate steppe 1.2 1 227 [187, 4 [3, 10]
Brassica months Germany, and broadleaf 308]
napus and cycle). Canada forest and taiga
B. Rotated with
campestris other crops
Cotton Annual (~6 0.3–0.4 China, Subtropical 1.2 35 299 [234, 10 [7, 12]
Gossypium months India monsoon, dry and 347]
hirsutum cycle). humid forest and
Rotated with temperate areas
other crops
Groundnuts Annual (4-5 0.5–0.8 China, Subtropical 1.5 6 351 [308, 11 [7, 16]
or peanuts months crop India monsoon, dry and 426]
Arachis cycle). humid forest and
hypogaea Rotated with temperate areas
other crops
Sunflower Annual (3-4 0.5–0.9 Ukraine, Temperate steppe 1.0 1 189 [177, 3 [2, 9]
Helianthus months crop Russia and broadleaf 222]
annuus cycle). forest
Rotated with
other crops
Coconut Perennial (30 0.4–2.4 Philippines, Tropical and n/a 65 317 [264, 73 [35,
Cocos – 50 y cycle) Indonesia, subtropical forest 414] 113]
nucifera India
Maize Annual (5-6 0.1–0.2 USA, China, Temperate steppe 0.7 131 273 [222, 9 [5, 20]
Zea mays months crop and broadleaf 427]
cycle). forest
Rotated with
other crops
Olive Perennial, 0.3–2.9 Spain, Italy, Mediterranean n/a 14 n/a n/a
Olea long lived. Greece vegetation
europaea Sometimes
inter-cropped
717