- Email: [email protected]
Appropriation of potential net primary production by cropland in terrestrial ecoregions
Journal of Cleaner Production 150 (2017) 294e300
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Appropriation of potential net primary production by cropland in terrestrial ecoregions , David Va r, Jan Weinzettel* Helena Medkova cka Charles University, Environment Centre, Jos e Martího 146/2, Prague 6, Czechia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 March 2016 Received in revised form 20 January 2017 Accepted 1 March 2017 Available online 2 March 2017
Terrestrial ecoregions of the world have been extensively converted to croplands as a result of human demand for food, fibers, fuels and fodder. Agricultural land cover change has been listed as one of the main drivers of biodiversity loss and change in ecosystem biogeochemical budgets. To provide a quantitative estimate of human impacts on ecosystems, we estimate the amount of net primary production appropriated by the world's croplands from potential natural vegetation cover. Potential net primary production embodied in the 170 crops analyzed was determined using a combination of existing spatial data on crop production and yield statistics, distribution of terrestrial ecoregions and net primary production of potential vegetation. We found that global croplands directly appropriate 9.1 Gt of carbon annually, which is about 14% of potential net primary production. The intensity of human impacts on terrestrial ecoregions differs according to the level of anthropogenic conversion to croplands. Temperate grasslands and savannas have been traditionally converted into croplands and therefore productivity appropriation reaches the highest levels of 34% in aggregate. In addition to the analysis of appropriation of net primary productivity in terrestrial regions, we also report intensity factors describing the embodied amount of productivity lost by conversion of natural regions into croplands. The results of appropriation of original natural net primary production by croplands contribute to discussion of the differences in land use intensity in different countries. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Sustainability Ecosystems Potential net primary production Appropriation Agriculture Cropland
1. Introduction Humans convert natural ecosystems to cropland to harness food, fodder, fibers and fuels. Land use change induced by the demand for crops has been accelerating in recent decades (Gibbs et al., 2010). The majority of this conversion has been located in biomes and ecoregions suitable for agriculture, and in tropical areas (Hoekstra et al., 2005). Human transformation of the biosphere has reached an unprecedented extent and rate. With the majority of natural biomes transformed, several planetary boundaries have €m been transgressed by human actions (Ellis et al., 2010; Rockstro et al., 2009). Vegetation intactness as a measure of ecosystem integrity in terrestrial ecoregions has been suggested as a key component to assess future vulnerability to climate change and habitat modification (Watson et al., 2013). Intact ecosystems provide valuable regulating ecosystem services which are further degraded as a result of land use change, especially in tropical
* Corresponding author. E-mail address: [email protected] (J. Weinzettel). http://dx.doi.org/10.1016/j.jclepro.2017.03.002 0959-6526/© 2017 Elsevier Ltd. All rights reserved.
regions (Costanza et al., 2014). Land transformation induced by agricultural crops has been the predominant driver of natural habitat loss. People have been converting productive ecosystems into croplands since the start of the Neolithic period, and the onset of farming has been proposed as one of the possible Anthropocene boundaries (Lewis and Maslin, 2015). In 2000, cropland area was estimated at 15 million km2 (Ramankutty et al., 2008), with only a minor change up to 2013 according to FAOSTAT (FAO, 2015). Net primary productivity (NPP) is the amount of biomass generated in an ecosystem over a particular time period. It has been frequently used as an ecological outcome of environmental processes (Costanza et al., 2007; Haberl et al., 2014). Through photosynthesis as one of the basic biosphere's energy flows, net primary production determines the amount of biomass and carbon available in ecosystems for humans and other species. The quantity of energy available in an ecosystem is also considered to be one of the major determinants of species diversity, especially species richness (Bailey et al., 2004). Human appropriation of net primary productivity (HANPP) has been used as a measure of various human
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
Abbreviations i j k A HA NPP C NPP0 P Y eNPP0 eNPP0f FCU
Cell identifier within the grid Crop identifier Country index Cropland area (measured in hectares) Harvested area (measured in hectares) Net primary production (measured in tons of carbon per year) Carbon Potential net primary production (measured in tons of carbon per year) Harvest (production) (measured in tons) Yield (Y ¼ P/HA) NPP0 embodied in a product or crop (measured in tons of carbon per year) NPP0 per ton of crop (measured in tons of carbon per year per ton of crop) Factor of cropland utilization (considers successive cropping and fallow land)
impacts on ecosystem integrity, through the analysis of NPP influenced by humans or remaining in ecosystems after human intervention (Allred et al., 2015; Haberl et al., 2014). The productivity of potential vegetation has been suggested as a proxy for ecosystem capacity to provide energy to ecosystems and goods and services supporting human society (Millennium Ecosystem Assessment, 2005). Previous studies of human appropriation of net primary production have focused on general patterns in the effects of land conversion and harvest on net primary productivity (Haberl et al., 2014). This approach enables differentiation of the effect of land cover change on actual net primary productivity and the effects of harvesting and biomass extraction from ecosystems. In this analysis, we quantify the direct appropriation of net primary production of potential vegetation by croplands. The conversion of land for productive purposes has been documented as the predominant driver of natural ecosystem productivity loss (DeFries, 2002; Haberl et al., 2014; Smith et al., 2014). Our approach is therefore different from previous studies of HANPP, which account for land conversion and harvesting (Haberl et al., 2007) or allocate NPP supply based on human demand (Imhoff et al., 2004). In this article, we present an analysis of appropriation of the potential productivity of natural ecosystems by global croplands. We focus on appropriation in the terrestrial ecoregions of the world as they represent relatively homogeneous natural units with regard to natural conditions, and hence net primary productivity patterns. We analyze the appropriation of potential net primary productivity embodied in 170 individual crops harvested on the site of the original vegetation. Following this analysis, we estimate factors which enable quantification of embodied net primary productivity in individual crops produced in world ecoregions and countries, respectively. These factors are constructed to assess the impact of crop production on natural ecosystems and can be applied in various analyses of the human transformation of Earth's land surface connected to crop production. This work is an essential step in consumption based accounting for appropriation of potential NPP.
2. Methods We define appropriation of potential net primary production
295
(NPP0) by cropland as the NPP0 of the natural area actually altered by cropland. We allocate this appropriation to crop products directly harvested from cropland (such as wheat or rice) to derive NPP0 embodied in those products (eNPP0, also called NPP0 footprint), based on the actual cropland area occupied by the crop and NPP0 of this area. Using spatially specific data and assuming homogenous spatial units, eNPP0 can be expressed in grid cell i for crop j as:
eNPP0i;j ¼ NPP00 i $Ai;j where i is a cell identifier within the grid, j is a crop identifier, NPP0i’ is NPP0 per unit of area and Ai,j is the actual cropland area allocated to crop j in cell i. In accordance with NPP0, eNPP0 is reported in the amount of carbon (e.g. tons) per year. In order to estimate the eNPP0 of individual crops it is necessary to derive Ai,j. As a hectare of cropland can be harvested more times a year (successive or multiple cropping), it is necessary to divide it among the harvested crops and therefore decrease the cropland area assigned to each crop accordingly. Conversely, cropland can also remain unharvested, which increases the cropland area with no additional harvest. We allocate the unharvested area to the harvested crops proportionally to their harvested area and therefore increase their assigned cropland area. In order to incorporate successive cropping and unharvested area to the eNPP0 calculation, we introduce a factor of cropland utilization (FCU) for adjustment of the rate of utilization of land use specific to each grid cell i. We calculate this factor by dividing the harvested area of all crops harvested in spatial unit i by the total cropland in this spatial unit:
P FCUi ¼
HAi;j
j
Ai
where HAi,j is the harvested area of crop j in cell i and Ai is the total cropland in cell i. This expression reflects the data availability, as harvested area is available for each crop in each cell and counts the total harvested area each time it is harvested (therefore, for multiple cropping it counts the harvested area multiple times), while the total cropland is available only as an aggregate over all crops and counts the cropland only once. We estimate cropland by crop and cell Ai,j from the crop and cell specific harvested area and cell specific factor of cropland utilization as:
Ai;j ¼
HAi;j FCUi
Hence, eNPP0i,j can be expressed as:
eNPP0i;j ¼
NPP00 i $HAi;j NPP00 i $Pi;j ¼ FCUi;j Yi;j $FCUi;j
(1)
After substituting
HAi;j ¼
Pi;j Yi;j
where Pi,j is the actual harvest of crop j in cell i and Yi,j is the yield of crop j in cell i. By summing eNPP0i,j over all spatial units within a defined area we obtain the total eNPP0j attributed to crop j within this area. The defined area can be an eco-region, country, or any other area of interest, e.g. a combination e an ecoregion within a country:
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
296
eNPP0j ¼
X
harvested, thus areas under successive or multiple cropping are counted more times. To identify the amount of net primary production appropriated by cropland area, it is important to account for real agricultural land use, and to count multiple cropped areas only once. Therefore, the area harvested has to be complemented by the data on cropland area with spatial distribution provided by Ramankutty et al. (2008), where each cropland area is counted only once. As the area of cropland (respectively ‘‘Arable lands and permanent crops’’ in FAOSTAT terminology) includes land temporarily fallow and unharvested, the amount of total global cropland exceeds the total area actually occupied by crops. Yet as fallow land originates from agriculture, we assign this excess to all crops proportionally to their harvested area as stated above. For potential net primary production, we used an existing layer developed by Haberl et al. (2007). Their calculation was based on atmospheric CO2 concentrations, gridded historical monthly climate data, and a soil-type classification at 0.5 spatial resolution. They used a 5 year average of the results from 1998 to 2002 for this dataset and resampled to 5 min resolution (Haberl et al., 2007), which corresponds to the other datasets we used. As potential net primary production comprises one of the basic ecological measures, we used terrestrial ecoregions of world classification to derive consistent patterns of NPP0 appropriation. Terrestrial ecoregions of the world are a consistent classification of terrestrial ecosystems based on biogeographical characteristics (Olson et al., 2001). World ecoregions are categorized into 14 major biomes (Table 1) consisting of 867 detailed ecoregions with an average size of 150,000 km2. We report our results for the 14 major biomes only.
eNPP0i;j
i
Summing eNPP0i,j over country k and dividing by the total harvest within the country results in factors (denoted eNPP0fj,k) which inform us about average NPP0 attributed to each ton of harvest of the respective crop in the respective country.
P
eNPP0i;j eNPP0fj;k ¼ i2kP Pi;j i2k
It can be seen from Equation (1) that eNPP0f factors increase with increasing potential net primary productivity and decrease with yield and the factor of cropland utilization.
3. Data To derive NPP0 appropriated by specific crops in global ecoregions and countries, we combined the following spatial datasets: (a) a spatially and crop specific harvested area (Monfreda et al., 2008); (b) cropland distribution (Ramankutty et al., 2008), and (c) potential NPP0 provided by Haberl et al. (2007). All those datasets are in the same format, with a 5 min resolution grid and global coverage with representative values for circa 2000. To allow an update of existing spatial datasets by the most widely available agricultural statistics, we followed FAOSTAT's classification system, and where needed data were re-classified to match the definitions and classification of FAOSTAT. To derive results for eco-regions and countries we used the Terrestrial Ecoregions dataset published by The Nature Conservancy (Olson et al., 2001) and country borders, respectively. The database developed by Monfreda et al. (2008) contains harvested area and yield for 175 crops. They distributed detailed statistical data of the harvested area of each crop over a cropland map within the smallest administrative unit available. Statistical data available between 1997 and 2003 were averaged to get a single representative value for circa 2000 (Monfreda et al., 2008). None of the other available data sets of crop distribution, e.g. (Leff et al., 2004; You and Wood, 2006) cover a global scale and wide range of crops at the same time. According to the definition of harvested area (FAO, 2015; Monfreda et al., 2008), the area is counted as many times as it is
4. Results In total, global croplands directly appropriate 9.1 Gt of carbon (C) annually, which is 14% of potential net primary production (NPP0). The map in Fig. 1 shows NPP0 appropriated by cropland in a 5 min resolution grid. It identifies the most affected regions around the globe. Concerning the global distribution of NPP0 appropriation, the areas with larger NPP0 appropriation can be found in regions with high natural levels of NPP0 and a large concentration of croplands. In some terrestrial biomes, people appropriate a considerably large share of net primary productivity (Fig. 2). Temperate
Table 1 NPP0 appropriated by cropland in ecoregions. Global total and ecoregions with more than 25% appropriated are highlighted in bold.
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Ecoregion
% of NPP0 global area [mil. total km2]
NPP0 [GtC/ year]/km2
NPP0 abs [GtC/ % NPP0 approp. by year] cropland
abs. NPP0 app. by cropland [GtC/year]
Tropical and Subtropical Moist Broadleaf Forests Tropical and Subtropical Dry Broadleaf Forests Tropical and Subtropical Coniferous Forests Temperate Broadleaf and Mixed Forests Temperate Conifer Forests Boreal Forests/Taiga Tropical and Subtropical Grasslands, Savannas and Shrublands Temperate Grasslands, Savannas and Shrublands Flooded Grasslands and Savannas Montane Grasslands and Shrublands Tundra Mediterranean Forests, Woodlands and Scrub Deserts and Xeric Shrublands Mangroves World
29% 4.2%
20 3.8
942 719
19 2.7
12% 25%
2.2 0.7
0.7% 11% 3.2% 11% 20%
0.6 13 4.4 16 19
735 566 478 457 657
0.5 7.3 2.1 7.4 13
19% 28% 6.6% 1.7% 10%
0.1 2.0 0.1 0.1 1.2
6.8%
10
460
4.4
34%
1.5
1.0% 2.9% 2.6% 1.9% 5.5% 0.4%
1.1 5.2 11 3.3 28 0.3
574 364 150 386 129 744 479
0.6 1.9 1.7 1.3 3.6 0.3 65.1
9.3% 8.8% 0.02% 29% 12% 18% 14%
0.1 0.2 0.0004 0.4 0.4 0.05 9.1
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
297
Fig. 1. NPP0 appropriated by cropland. Average values per cell in the 5 min grid resolution.
grasslands and savannas, distributed across Eurasia, North and South America and Australia, have been traditionally converted into croplands and therefore the NPP0 appropriation there reaches the highest levels of 34% of NPP0 per year (1.5 Gt C). In originally forested biomes such as Mediterranean Woodlands, Temperate Forests, and Tropical and Subtropical Dry Broadleaved Forests, this proportion reaches 25%e29% of original net primary productivity (Table 1). In absolute terms, people appropriate most NPP0 in Tropical and Subtropical Moist Broadleaved Forest (2.2 Gt C). However, with regard to the large remaining area of the forest, relative NPP0 appropriation reaches 12%. In Temperate Broadleaf and Mixed Forests ecoregions, croplands appropriate 28% of NPP0, with a significant contribution to global NPP0 appropriation (11%). In other biomes, croplands appropriate a still significant part of ecoregion production, but with only a smaller contribution to total NPP0 appropriation. These include, for example, Tropical and Subtropical Dry Broadleaved Forest (25% of NPP0), Tropical and Subtropical Coniferous Forests (19%), Mangroves (18%) and several other ecoregions (Table 1). We also assessed NPP0 appropriation by particular crops. The most important crops dominating appropriation of NPP0 in terrestrial ecosystems worldwide are shown in Table 2. The first six crops (wheat, maize, rice, pumpkins for fodder, soybeans and barley) are responsible for more than half of NPP0 appropriated by cropland. The area currently occupied by the top ten crops contribute 60%, with the other more than 150 crops responsible only for 40% of the appropriated NPP0. In Table 3 we focus more on the four most utilized ecoregions and report the top five crops. It can be seen that maize and wheat
are among the top 5 crops in all those eco-regions, followed by barley and soybeans. The top 5 crops account for more than half of the appropriated NPP0 in all those ecoregions. In the next section we determine the amount of carbon which individual crops appropriate in different countries per ton of harvest. Proposed eNPP0f factors reflect natural conditions as well as agricultural practices - amount of yield and rate of cropland utilization (FCU). Following these components, we can distinguish areas where most productive natural biomes are exposed to intense agriculture and determine the crops which cause the highest appropriation, as well as detect areas with less efficient agriculture. We present factors for the top ten crops from Table 1 and countries where cropland appropriates more than 0.4 Gt C annually (Table 4). For example, in the United States the wheat harvest appropriates 2.8 tons of natural NPP0 per 1 ton of harvest, while in India and China it is 1.2e1.3 tons of NPP0 per ton of harvest (due to high yield and/or low NPP0) and in Russia 4.3 tons of NPP0 per ton of harvest (low yield and low FCU). Similar patterns can be found for other crops; generally, the United States have mean or low factor values due to relatively high yields, while low yields in Russia and India cause higher factor values. High appropriation of NPP0 is also caused by low FCU in Russia (with a lot of unharvested area and fallow land) or very high NPP0 in Indonesia. Low values of eNPP0f for sugarcane reflect very high yields (20e30 higher yield than wheat due to lower dry fraction). High values for the crop “Beans, dry” can be explained by a very low yield (India) or by a combination with very high NPP0 and/or low FCU (Indonesia, Brazil). The complete dataset is available in the online SI. Concerning world-wide production, the highest appropriation of NPP0 per one ton of harvest is achieved by spices (e.g. vanilla 297,
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
298
Fig. 2. Eco-regions by NPP0 appropriated by cropland.
Table 2 Top ten crops in terms of NPP0 appropriation. FAO code
Crop name FAO
Share of NPP0 appr. by cropland
Cumulative share
15 56 27 645
Wheat Maize Rice, paddy Pumpkins for Fodder Soybeans Barley Sorghum Millet Beans, dry Sugar cane Rest
13% 12% 11% 6%
13% 25% 35% 41%
6% 4% 3% 2% 2% 2% 40%
47% 51% 54% 56% 58% 60% 100%
236 44 83 79 176 156
cloves 140 or cinnamon 35 tons of carbon per ton of harvest). High appropriating crops also include e.g. cocoa beans with 26 tons of NPP0 per ton of harvest or green coffee and sesame seeds with 20 tons of NPP0 per ton of harvest. The lowest appropriation belongs to crops with low dry fraction (sugar cane 0.15; watermelons 0.2), crops with a short growing season (lettuce 0.22) and/or mass indoor production (mushrooms 0.03). In order to provide a broader socio-ecological perspective, we also focus on the country level. When an amount of NPP0 appropriated by a country's total production is compared to potential appropriation by the same production with world mean crop factors, most of Africa appropriates considerably higher NPP0 for its production (Zambia 11 more than the world mean, Botswana 5 more). Other high appropriating countries are Mongolia (18 more), Brunei (12 more) or Iraq (10 more). Latin America, South
East Asia, East Europe and Russia achieve a significantly higher appropriation of NPP0 per ton of harvest than West Europe, South Asia and North America. Countries widely utilizing irrigation in areas with low NPP0 (Egypt or Saudi Arabia) have the lowest values of NPP0f. 5. Discussion People are influencing, or appropriating, net primary production by several means. First, human activities transform land cover and change actual land use. Agricultural practices can even lead to increases in NPP in originally non-forested areas, but only when receiving external outputs, e.g. irrigation or fertilization (DeFries, 2002; Smith et al., 2014). Natural vegetation can be degraded with extensive NPP reduction, especially in tropical and subtropical regions. Urbanization and other land development effects are components of profound land transformation induced by a range of human activities (Allred et al., 2015; Imhoff et al., 2004). Second, people harvest food, fodder, fibers, fuels, timber and other materials from ecosystems. These direct removals of biomass are usually a dominant factor in NPP appropriation (Haberl et al., 2007). Lastly, humans influence NPP by other processes like air pollution, human-induced fires or soil erosion on agricultural land (Haberl et al., 2007; Vitousek et al., 1986). Factors of eNPP0f defined in this study reflect potential net primary productivity as well as agricultural practices. High factor values are achieved by areas where most productive natural ecosystems are exposed to intense agriculture (Indonesia) or areas with low efficient agriculture where yields are low and a considerable part of the land is fallow or unharvested (Russia). Areas with low NPP0 and extensive irrigation (Egypt, Saudi Arabia) have the
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
299
Table 3 NPP0 appropriated by cropland by ecoregion and top 5 crops. The repeating crops are highlighted by the same color.
Table 4 NPP0 in tons of carbon per ton of harvest.
lowest factor values which cannot be interpreted as low pressure on the environment due to the high consumption of scarce water in dry regions. Irrigation causes serious environmental threats for soil and water supplies not only in arid areas, and is not sustainable in the long-term. The differences in eNPP0f factors pose the question whether international trade can in principal contribute to decreasing NPP0 appropriated by cropland by shifting crops from more effective countries to less. However, international trade has been shown as a medium by which rich countries can extend their footprint beyond their boundaries by Weinzettel et al. (2013) and the current state of international trade is driven by the desire for price minimization and not the minimization of environmental burden. Furthermore, such an analysis would require data on the marginal NPP0f which are expected to play the most important role in such a process. The eNPP0f factors and their determinants could also be used to evaluate the potential of technological development for the minimization of human appropriation of NPP0 as closing global yield gaps has been suggested as one of the challenges for food security and effective agricultural production (Lobell et al., 2009). However, both those analyses are beyond the scope of this paper. In this analysis, we assess only the level of appropriation of potential productivity in terrestrial ecoregions posed by the harvesting of crops. For example, mangroves form a biome endangered by many other activities such as clearing for coastal development or shrimp aquaculture. Those biomes not suitable for agriculture have a low share of NPP0 appropriated by croplands (18% of natural potential productivity), although the appropriation of NPP and ecosystem burden can take place by other means. NPP0 have been often used as a reference for actual NPP in land use intensity studies which show that crop production especially in tropical regions is
considerably reduced in comparison to the potential productivity (Niedertscheider et al., 2016). Our results complement these findings that in terrestrial ecoregions, cropland production embodies a significant fraction of NPP0. However, the land use intensity measured by comparison of actual and potential NPP does not reveal the overall outcome of human appropriation of ecosystems. Only the tundra ecoregion can be considered relatively untouched by agriculture, followed by boreal forests (taiga). Otherwise, people have appropriated even parts of desert and xeric ecoregions. 6. Conclusions The presented analysis focuses on the extent of anthropogenic transformation of world ecoregions by cropland. We document the level of the transformation of terrestrial ecoregions by the extent of appropriation of potential net natural productivity (NPP0). Comparing terrestrial ecoregions, Temperate Grasslands, Savannas and Shrublands is the most exposed biome with 34% of NPP0 appropriated by cropland. However, in absolute terms, people appropriate most of NPP0 in Tropical and Subtropical Moist Broadleaved Forest (2.2 Gt C). The first six crops by total amount of appropriation of NPP0 (wheat, maize, rice, pumpkins for fodder, soybeans and barley) are responsible for more than half of total NPP0 appropriated by cropland world-wide (9.1 Gt C annually). Replacement of natural vegetation by croplands significantly influences the structure and functioning of habitats, contributing to biodiversity loss. Our analysis develops a novel approach to account for the transformation of ecosystems by croplands. Potential NPP footprint, introduced in this analysis, quantifies NPP of potential vegetation in terrestrial ecoregions, embodied in agricultural products that replace the original vegetation. We found that
300
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
globally, people appropriate 14% of potential net primary productivity by agriculture, but with large regional differences. In traditionally exploited regions, croplands appropriate more than 30% of potential NPP. The factors of NPP0 appropriation we developed reflect natural productivity as well as agricultural practices. Therefore, the results reported in this analysis can be used in strengthening strategies to alleviate agriculture e nature conservation conflicts in terrestrial ecoregions of the world. These findings can support strategies and policies related to food security, nature friendly agricultural practices and biodiversity conservation within planetary boundaries. Acknowledgement This work has been supported by the Czech Science Foundation via grant number 14-05292S. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jclepro.2017.03.002. References Allred, B.W., Smith, W.K., Twidwell, D., Haggerty, J.H., Running, S.W., Naugle, D.E., Fuhlendorf, S.D., 2015. Ecosystem services lost to oil and gas in North America. Science 348, 401e402. Bailey, S.A., Horner-Devine, M.C., Luck, G., Moore, L.A., Carney, K.M., Anderson, S., Betrus, C., Fleishman, E., 2004. Primary productivity and species richness: relationships among functional guilds, residency groups and vagility classes at multiple spatial scales. Ecography 27, 207e217. Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S.J., Kubiszewski, I., Farber, S., Turner, R.K., 2014. Changes in the global value of ecosystem services. Glob. Environ. Change 26, 152e158. Costanza, R., Fisher, B., Mulder, K., Liu, S., Christopher, T., 2007. Biodiversity and ecosystem services: a multi-scale empirical study of the relationship between species richness and net primary production. Ecol. Econ. 61, 478e491. DeFries, R., 2002. Past and future sensitivity of primary production to human modification of the landscape. Geophys. Res. Lett. 29, 31e36. Ellis, E.C., Goldewijk, K.K., Siebert, S., Lightman, D., Ramankutty, N., 2010. Anthropogenic transformation of the biomes, 1700 to 2000. Glob. Ecol. Biogeogr. 19, 589e606. FAO, 2015. FAOSTAT Database. Food and Agriculture Organization of the United Nations, Rome, Italy. Gibbs, H.K., Ruesch, A.S., Achard, F., Clayton, M.K., Holmgren, P., Ramankutty, N.,
Foley, J.A., 2010. Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. Proc. Natl. Acad. Sci. U. S. A. 107, 16732e16737. Haberl, H., Erb, K.H., Krausmann, F., 2014. Human appropriation of net primary production: patterns, trends, and planetary boundaries. Annu. Rev. Environ. Resour. 363e391. Haberl, H., Erb, K.H., Krausmann, F., Gaube, V., Bondeau, A., Plutzar, C., Gingrich, S., Lucht, W., Fischer-Kowalski, M., 2007. Quantifying and mapping the human appropriation of net primary production in earth's terrestrial ecosystems. Proc. Natl. Acad. Sci. U. S. A. 104, 12942e12945. Hoekstra, J.M., Boucher, T.M., Ricketts, T.H., Roberts, C., 2005. Confronting a biome crisis: global disparities of habitat loss and protection. Ecol. Lett. 8, 23e29. Imhoff, M.L., Bounoua, L., Ricketts, T., Loucks, C., Harriss, R., Lawrence, W.T., 2004. Global patterns in human consumption of net primary production. Nature 429, 870e873. Leff, B., Ramankutty, N., Foley, J.A., 2004. Geographic distribution of major crops across the world. Glob. Biogeochem. Cycles 18, 1001e1027. GB1009. Lewis, S.L., Maslin, M.A., 2015. Defining the Anthropocene. Nature 519, 171e180. Lobell, D.B., Cassman, K.G., Field, C.B., 2009. Crop yield gaps: their importance, magnitudes, and causes. Annu. Rev. Environ. Resour. 179e204. Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-being: Synthesis. Island Press, Washington, DC. Monfreda, C., Ramankutty, N., Foley, J.A., 2008. Farming the planet: 2. Geographic distribution of crop areas, yields, physiological types, and net primary production in the year 2000. Glob. Biogeochem. Cycles 22. Niedertscheider, M., Kastner, T., Fetzel, T., Haberl, H., Kroisleitner, C., Plutzar, C., Erb, K.H., 2016. Mapping and analysing cropland use intensity from a NPP perspective. Environ. Res. Lett. 11. Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V.N., Underwood, E.C., D'Amico, J.A., Itoua, I., Strand, H.E., Morrison, J.C., Loucks, C.J., Allnutt, T.F., Ricketts, T.H., Kura, Y., Lamoreux, J.F., Wettengel, W.W., Hedao, P., Kassem, K.R., 2001. Terrestrial ecoregions of the world: a new map of life on Earth. Bioscience 51, 933e938. Ramankutty, N., Evan, A.T., Monfreda, C., Foley, J.A., 2008. Farming the planet: 1. Geographic distribution of global agricultural lands in the year 2000. Glob. Biogeochem. Cycles 22. €m, J., Steffen, W., Noone, K., Persson, Å., Chapin, F.S., Lambin, E.F., Rockstro Lenton, T.M., Scheffer, M., Folke, C., Schellnhuber, H.J., Nykvist, B., De Wit, C.A., €rlin, S., Snyder, P.K., Costanza, R., Hughes, T., Van Der Leeuw, S., Rodhe, H., So Svedin, U., Falkenmark, M., Karlberg, L., Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P., Foley, J.A., 2009. A safe operating space for humanity. Nature 461, 472e475. Smith, W.K., Cleveland, C.C., Reed, S.C., Running, S.W., 2014. Agricultural conversion without external water and nutrient inputs reduces terrestrial vegetation productivity. Geophys. Res. Lett. 41, 449e455. Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., Matson, P.A., 1986. Human appropriation of the products of photosynthesis. Bioscience 36, 368e373. Watson, J.E.M., Iwamura, T., Butt, N., 2013. Mapping vulnerability and conservation adaptation strategies under climate change. Nat. Clim. Change 3, 989e994. Weinzettel, J., Hertwich, E.G., Peters, G.P., Steen-Olsen, K., Galli, A., 2013. Affluence drives the global displacement of land use. Glob. Environ. Change 23, 433e438. You, L., Wood, S., 2006. An entropy approach to spatial disaggregation of agricultural production. Agric. Syst. 90, 329e347.
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Appropriation of potential net primary production by cropland in terrestrial ecoregions , David Va r, Jan Weinzettel* Helena Medkova cka Charles University, Environment Centre, Jos e Martího 146/2, Prague 6, Czechia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 March 2016 Received in revised form 20 January 2017 Accepted 1 March 2017 Available online 2 March 2017
Terrestrial ecoregions of the world have been extensively converted to croplands as a result of human demand for food, fibers, fuels and fodder. Agricultural land cover change has been listed as one of the main drivers of biodiversity loss and change in ecosystem biogeochemical budgets. To provide a quantitative estimate of human impacts on ecosystems, we estimate the amount of net primary production appropriated by the world's croplands from potential natural vegetation cover. Potential net primary production embodied in the 170 crops analyzed was determined using a combination of existing spatial data on crop production and yield statistics, distribution of terrestrial ecoregions and net primary production of potential vegetation. We found that global croplands directly appropriate 9.1 Gt of carbon annually, which is about 14% of potential net primary production. The intensity of human impacts on terrestrial ecoregions differs according to the level of anthropogenic conversion to croplands. Temperate grasslands and savannas have been traditionally converted into croplands and therefore productivity appropriation reaches the highest levels of 34% in aggregate. In addition to the analysis of appropriation of net primary productivity in terrestrial regions, we also report intensity factors describing the embodied amount of productivity lost by conversion of natural regions into croplands. The results of appropriation of original natural net primary production by croplands contribute to discussion of the differences in land use intensity in different countries. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Sustainability Ecosystems Potential net primary production Appropriation Agriculture Cropland
1. Introduction Humans convert natural ecosystems to cropland to harness food, fodder, fibers and fuels. Land use change induced by the demand for crops has been accelerating in recent decades (Gibbs et al., 2010). The majority of this conversion has been located in biomes and ecoregions suitable for agriculture, and in tropical areas (Hoekstra et al., 2005). Human transformation of the biosphere has reached an unprecedented extent and rate. With the majority of natural biomes transformed, several planetary boundaries have €m been transgressed by human actions (Ellis et al., 2010; Rockstro et al., 2009). Vegetation intactness as a measure of ecosystem integrity in terrestrial ecoregions has been suggested as a key component to assess future vulnerability to climate change and habitat modification (Watson et al., 2013). Intact ecosystems provide valuable regulating ecosystem services which are further degraded as a result of land use change, especially in tropical
* Corresponding author. E-mail address: [email protected] (J. Weinzettel). http://dx.doi.org/10.1016/j.jclepro.2017.03.002 0959-6526/© 2017 Elsevier Ltd. All rights reserved.
regions (Costanza et al., 2014). Land transformation induced by agricultural crops has been the predominant driver of natural habitat loss. People have been converting productive ecosystems into croplands since the start of the Neolithic period, and the onset of farming has been proposed as one of the possible Anthropocene boundaries (Lewis and Maslin, 2015). In 2000, cropland area was estimated at 15 million km2 (Ramankutty et al., 2008), with only a minor change up to 2013 according to FAOSTAT (FAO, 2015). Net primary productivity (NPP) is the amount of biomass generated in an ecosystem over a particular time period. It has been frequently used as an ecological outcome of environmental processes (Costanza et al., 2007; Haberl et al., 2014). Through photosynthesis as one of the basic biosphere's energy flows, net primary production determines the amount of biomass and carbon available in ecosystems for humans and other species. The quantity of energy available in an ecosystem is also considered to be one of the major determinants of species diversity, especially species richness (Bailey et al., 2004). Human appropriation of net primary productivity (HANPP) has been used as a measure of various human
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
Abbreviations i j k A HA NPP C NPP0 P Y eNPP0 eNPP0f FCU
Cell identifier within the grid Crop identifier Country index Cropland area (measured in hectares) Harvested area (measured in hectares) Net primary production (measured in tons of carbon per year) Carbon Potential net primary production (measured in tons of carbon per year) Harvest (production) (measured in tons) Yield (Y ¼ P/HA) NPP0 embodied in a product or crop (measured in tons of carbon per year) NPP0 per ton of crop (measured in tons of carbon per year per ton of crop) Factor of cropland utilization (considers successive cropping and fallow land)
impacts on ecosystem integrity, through the analysis of NPP influenced by humans or remaining in ecosystems after human intervention (Allred et al., 2015; Haberl et al., 2014). The productivity of potential vegetation has been suggested as a proxy for ecosystem capacity to provide energy to ecosystems and goods and services supporting human society (Millennium Ecosystem Assessment, 2005). Previous studies of human appropriation of net primary production have focused on general patterns in the effects of land conversion and harvest on net primary productivity (Haberl et al., 2014). This approach enables differentiation of the effect of land cover change on actual net primary productivity and the effects of harvesting and biomass extraction from ecosystems. In this analysis, we quantify the direct appropriation of net primary production of potential vegetation by croplands. The conversion of land for productive purposes has been documented as the predominant driver of natural ecosystem productivity loss (DeFries, 2002; Haberl et al., 2014; Smith et al., 2014). Our approach is therefore different from previous studies of HANPP, which account for land conversion and harvesting (Haberl et al., 2007) or allocate NPP supply based on human demand (Imhoff et al., 2004). In this article, we present an analysis of appropriation of the potential productivity of natural ecosystems by global croplands. We focus on appropriation in the terrestrial ecoregions of the world as they represent relatively homogeneous natural units with regard to natural conditions, and hence net primary productivity patterns. We analyze the appropriation of potential net primary productivity embodied in 170 individual crops harvested on the site of the original vegetation. Following this analysis, we estimate factors which enable quantification of embodied net primary productivity in individual crops produced in world ecoregions and countries, respectively. These factors are constructed to assess the impact of crop production on natural ecosystems and can be applied in various analyses of the human transformation of Earth's land surface connected to crop production. This work is an essential step in consumption based accounting for appropriation of potential NPP.
2. Methods We define appropriation of potential net primary production
295
(NPP0) by cropland as the NPP0 of the natural area actually altered by cropland. We allocate this appropriation to crop products directly harvested from cropland (such as wheat or rice) to derive NPP0 embodied in those products (eNPP0, also called NPP0 footprint), based on the actual cropland area occupied by the crop and NPP0 of this area. Using spatially specific data and assuming homogenous spatial units, eNPP0 can be expressed in grid cell i for crop j as:
eNPP0i;j ¼ NPP00 i $Ai;j where i is a cell identifier within the grid, j is a crop identifier, NPP0i’ is NPP0 per unit of area and Ai,j is the actual cropland area allocated to crop j in cell i. In accordance with NPP0, eNPP0 is reported in the amount of carbon (e.g. tons) per year. In order to estimate the eNPP0 of individual crops it is necessary to derive Ai,j. As a hectare of cropland can be harvested more times a year (successive or multiple cropping), it is necessary to divide it among the harvested crops and therefore decrease the cropland area assigned to each crop accordingly. Conversely, cropland can also remain unharvested, which increases the cropland area with no additional harvest. We allocate the unharvested area to the harvested crops proportionally to their harvested area and therefore increase their assigned cropland area. In order to incorporate successive cropping and unharvested area to the eNPP0 calculation, we introduce a factor of cropland utilization (FCU) for adjustment of the rate of utilization of land use specific to each grid cell i. We calculate this factor by dividing the harvested area of all crops harvested in spatial unit i by the total cropland in this spatial unit:
P FCUi ¼
HAi;j
j
Ai
where HAi,j is the harvested area of crop j in cell i and Ai is the total cropland in cell i. This expression reflects the data availability, as harvested area is available for each crop in each cell and counts the total harvested area each time it is harvested (therefore, for multiple cropping it counts the harvested area multiple times), while the total cropland is available only as an aggregate over all crops and counts the cropland only once. We estimate cropland by crop and cell Ai,j from the crop and cell specific harvested area and cell specific factor of cropland utilization as:
Ai;j ¼
HAi;j FCUi
Hence, eNPP0i,j can be expressed as:
eNPP0i;j ¼
NPP00 i $HAi;j NPP00 i $Pi;j ¼ FCUi;j Yi;j $FCUi;j
(1)
After substituting
HAi;j ¼
Pi;j Yi;j
where Pi,j is the actual harvest of crop j in cell i and Yi,j is the yield of crop j in cell i. By summing eNPP0i,j over all spatial units within a defined area we obtain the total eNPP0j attributed to crop j within this area. The defined area can be an eco-region, country, or any other area of interest, e.g. a combination e an ecoregion within a country:
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
296
eNPP0j ¼
X
harvested, thus areas under successive or multiple cropping are counted more times. To identify the amount of net primary production appropriated by cropland area, it is important to account for real agricultural land use, and to count multiple cropped areas only once. Therefore, the area harvested has to be complemented by the data on cropland area with spatial distribution provided by Ramankutty et al. (2008), where each cropland area is counted only once. As the area of cropland (respectively ‘‘Arable lands and permanent crops’’ in FAOSTAT terminology) includes land temporarily fallow and unharvested, the amount of total global cropland exceeds the total area actually occupied by crops. Yet as fallow land originates from agriculture, we assign this excess to all crops proportionally to their harvested area as stated above. For potential net primary production, we used an existing layer developed by Haberl et al. (2007). Their calculation was based on atmospheric CO2 concentrations, gridded historical monthly climate data, and a soil-type classification at 0.5 spatial resolution. They used a 5 year average of the results from 1998 to 2002 for this dataset and resampled to 5 min resolution (Haberl et al., 2007), which corresponds to the other datasets we used. As potential net primary production comprises one of the basic ecological measures, we used terrestrial ecoregions of world classification to derive consistent patterns of NPP0 appropriation. Terrestrial ecoregions of the world are a consistent classification of terrestrial ecosystems based on biogeographical characteristics (Olson et al., 2001). World ecoregions are categorized into 14 major biomes (Table 1) consisting of 867 detailed ecoregions with an average size of 150,000 km2. We report our results for the 14 major biomes only.
eNPP0i;j
i
Summing eNPP0i,j over country k and dividing by the total harvest within the country results in factors (denoted eNPP0fj,k) which inform us about average NPP0 attributed to each ton of harvest of the respective crop in the respective country.
P
eNPP0i;j eNPP0fj;k ¼ i2kP Pi;j i2k
It can be seen from Equation (1) that eNPP0f factors increase with increasing potential net primary productivity and decrease with yield and the factor of cropland utilization.
3. Data To derive NPP0 appropriated by specific crops in global ecoregions and countries, we combined the following spatial datasets: (a) a spatially and crop specific harvested area (Monfreda et al., 2008); (b) cropland distribution (Ramankutty et al., 2008), and (c) potential NPP0 provided by Haberl et al. (2007). All those datasets are in the same format, with a 5 min resolution grid and global coverage with representative values for circa 2000. To allow an update of existing spatial datasets by the most widely available agricultural statistics, we followed FAOSTAT's classification system, and where needed data were re-classified to match the definitions and classification of FAOSTAT. To derive results for eco-regions and countries we used the Terrestrial Ecoregions dataset published by The Nature Conservancy (Olson et al., 2001) and country borders, respectively. The database developed by Monfreda et al. (2008) contains harvested area and yield for 175 crops. They distributed detailed statistical data of the harvested area of each crop over a cropland map within the smallest administrative unit available. Statistical data available between 1997 and 2003 were averaged to get a single representative value for circa 2000 (Monfreda et al., 2008). None of the other available data sets of crop distribution, e.g. (Leff et al., 2004; You and Wood, 2006) cover a global scale and wide range of crops at the same time. According to the definition of harvested area (FAO, 2015; Monfreda et al., 2008), the area is counted as many times as it is
4. Results In total, global croplands directly appropriate 9.1 Gt of carbon (C) annually, which is 14% of potential net primary production (NPP0). The map in Fig. 1 shows NPP0 appropriated by cropland in a 5 min resolution grid. It identifies the most affected regions around the globe. Concerning the global distribution of NPP0 appropriation, the areas with larger NPP0 appropriation can be found in regions with high natural levels of NPP0 and a large concentration of croplands. In some terrestrial biomes, people appropriate a considerably large share of net primary productivity (Fig. 2). Temperate
Table 1 NPP0 appropriated by cropland in ecoregions. Global total and ecoregions with more than 25% appropriated are highlighted in bold.
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Ecoregion
% of NPP0 global area [mil. total km2]
NPP0 [GtC/ year]/km2
NPP0 abs [GtC/ % NPP0 approp. by year] cropland
abs. NPP0 app. by cropland [GtC/year]
Tropical and Subtropical Moist Broadleaf Forests Tropical and Subtropical Dry Broadleaf Forests Tropical and Subtropical Coniferous Forests Temperate Broadleaf and Mixed Forests Temperate Conifer Forests Boreal Forests/Taiga Tropical and Subtropical Grasslands, Savannas and Shrublands Temperate Grasslands, Savannas and Shrublands Flooded Grasslands and Savannas Montane Grasslands and Shrublands Tundra Mediterranean Forests, Woodlands and Scrub Deserts and Xeric Shrublands Mangroves World
29% 4.2%
20 3.8
942 719
19 2.7
12% 25%
2.2 0.7
0.7% 11% 3.2% 11% 20%
0.6 13 4.4 16 19
735 566 478 457 657
0.5 7.3 2.1 7.4 13
19% 28% 6.6% 1.7% 10%
0.1 2.0 0.1 0.1 1.2
6.8%
10
460
4.4
34%
1.5
1.0% 2.9% 2.6% 1.9% 5.5% 0.4%
1.1 5.2 11 3.3 28 0.3
574 364 150 386 129 744 479
0.6 1.9 1.7 1.3 3.6 0.3 65.1
9.3% 8.8% 0.02% 29% 12% 18% 14%
0.1 0.2 0.0004 0.4 0.4 0.05 9.1
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
297
Fig. 1. NPP0 appropriated by cropland. Average values per cell in the 5 min grid resolution.
grasslands and savannas, distributed across Eurasia, North and South America and Australia, have been traditionally converted into croplands and therefore the NPP0 appropriation there reaches the highest levels of 34% of NPP0 per year (1.5 Gt C). In originally forested biomes such as Mediterranean Woodlands, Temperate Forests, and Tropical and Subtropical Dry Broadleaved Forests, this proportion reaches 25%e29% of original net primary productivity (Table 1). In absolute terms, people appropriate most NPP0 in Tropical and Subtropical Moist Broadleaved Forest (2.2 Gt C). However, with regard to the large remaining area of the forest, relative NPP0 appropriation reaches 12%. In Temperate Broadleaf and Mixed Forests ecoregions, croplands appropriate 28% of NPP0, with a significant contribution to global NPP0 appropriation (11%). In other biomes, croplands appropriate a still significant part of ecoregion production, but with only a smaller contribution to total NPP0 appropriation. These include, for example, Tropical and Subtropical Dry Broadleaved Forest (25% of NPP0), Tropical and Subtropical Coniferous Forests (19%), Mangroves (18%) and several other ecoregions (Table 1). We also assessed NPP0 appropriation by particular crops. The most important crops dominating appropriation of NPP0 in terrestrial ecosystems worldwide are shown in Table 2. The first six crops (wheat, maize, rice, pumpkins for fodder, soybeans and barley) are responsible for more than half of NPP0 appropriated by cropland. The area currently occupied by the top ten crops contribute 60%, with the other more than 150 crops responsible only for 40% of the appropriated NPP0. In Table 3 we focus more on the four most utilized ecoregions and report the top five crops. It can be seen that maize and wheat
are among the top 5 crops in all those eco-regions, followed by barley and soybeans. The top 5 crops account for more than half of the appropriated NPP0 in all those ecoregions. In the next section we determine the amount of carbon which individual crops appropriate in different countries per ton of harvest. Proposed eNPP0f factors reflect natural conditions as well as agricultural practices - amount of yield and rate of cropland utilization (FCU). Following these components, we can distinguish areas where most productive natural biomes are exposed to intense agriculture and determine the crops which cause the highest appropriation, as well as detect areas with less efficient agriculture. We present factors for the top ten crops from Table 1 and countries where cropland appropriates more than 0.4 Gt C annually (Table 4). For example, in the United States the wheat harvest appropriates 2.8 tons of natural NPP0 per 1 ton of harvest, while in India and China it is 1.2e1.3 tons of NPP0 per ton of harvest (due to high yield and/or low NPP0) and in Russia 4.3 tons of NPP0 per ton of harvest (low yield and low FCU). Similar patterns can be found for other crops; generally, the United States have mean or low factor values due to relatively high yields, while low yields in Russia and India cause higher factor values. High appropriation of NPP0 is also caused by low FCU in Russia (with a lot of unharvested area and fallow land) or very high NPP0 in Indonesia. Low values of eNPP0f for sugarcane reflect very high yields (20e30 higher yield than wheat due to lower dry fraction). High values for the crop “Beans, dry” can be explained by a very low yield (India) or by a combination with very high NPP0 and/or low FCU (Indonesia, Brazil). The complete dataset is available in the online SI. Concerning world-wide production, the highest appropriation of NPP0 per one ton of harvest is achieved by spices (e.g. vanilla 297,
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
298
Fig. 2. Eco-regions by NPP0 appropriated by cropland.
Table 2 Top ten crops in terms of NPP0 appropriation. FAO code
Crop name FAO
Share of NPP0 appr. by cropland
Cumulative share
15 56 27 645
Wheat Maize Rice, paddy Pumpkins for Fodder Soybeans Barley Sorghum Millet Beans, dry Sugar cane Rest
13% 12% 11% 6%
13% 25% 35% 41%
6% 4% 3% 2% 2% 2% 40%
47% 51% 54% 56% 58% 60% 100%
236 44 83 79 176 156
cloves 140 or cinnamon 35 tons of carbon per ton of harvest). High appropriating crops also include e.g. cocoa beans with 26 tons of NPP0 per ton of harvest or green coffee and sesame seeds with 20 tons of NPP0 per ton of harvest. The lowest appropriation belongs to crops with low dry fraction (sugar cane 0.15; watermelons 0.2), crops with a short growing season (lettuce 0.22) and/or mass indoor production (mushrooms 0.03). In order to provide a broader socio-ecological perspective, we also focus on the country level. When an amount of NPP0 appropriated by a country's total production is compared to potential appropriation by the same production with world mean crop factors, most of Africa appropriates considerably higher NPP0 for its production (Zambia 11 more than the world mean, Botswana 5 more). Other high appropriating countries are Mongolia (18 more), Brunei (12 more) or Iraq (10 more). Latin America, South
East Asia, East Europe and Russia achieve a significantly higher appropriation of NPP0 per ton of harvest than West Europe, South Asia and North America. Countries widely utilizing irrigation in areas with low NPP0 (Egypt or Saudi Arabia) have the lowest values of NPP0f. 5. Discussion People are influencing, or appropriating, net primary production by several means. First, human activities transform land cover and change actual land use. Agricultural practices can even lead to increases in NPP in originally non-forested areas, but only when receiving external outputs, e.g. irrigation or fertilization (DeFries, 2002; Smith et al., 2014). Natural vegetation can be degraded with extensive NPP reduction, especially in tropical and subtropical regions. Urbanization and other land development effects are components of profound land transformation induced by a range of human activities (Allred et al., 2015; Imhoff et al., 2004). Second, people harvest food, fodder, fibers, fuels, timber and other materials from ecosystems. These direct removals of biomass are usually a dominant factor in NPP appropriation (Haberl et al., 2007). Lastly, humans influence NPP by other processes like air pollution, human-induced fires or soil erosion on agricultural land (Haberl et al., 2007; Vitousek et al., 1986). Factors of eNPP0f defined in this study reflect potential net primary productivity as well as agricultural practices. High factor values are achieved by areas where most productive natural ecosystems are exposed to intense agriculture (Indonesia) or areas with low efficient agriculture where yields are low and a considerable part of the land is fallow or unharvested (Russia). Areas with low NPP0 and extensive irrigation (Egypt, Saudi Arabia) have the
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
299
Table 3 NPP0 appropriated by cropland by ecoregion and top 5 crops. The repeating crops are highlighted by the same color.
Table 4 NPP0 in tons of carbon per ton of harvest.
lowest factor values which cannot be interpreted as low pressure on the environment due to the high consumption of scarce water in dry regions. Irrigation causes serious environmental threats for soil and water supplies not only in arid areas, and is not sustainable in the long-term. The differences in eNPP0f factors pose the question whether international trade can in principal contribute to decreasing NPP0 appropriated by cropland by shifting crops from more effective countries to less. However, international trade has been shown as a medium by which rich countries can extend their footprint beyond their boundaries by Weinzettel et al. (2013) and the current state of international trade is driven by the desire for price minimization and not the minimization of environmental burden. Furthermore, such an analysis would require data on the marginal NPP0f which are expected to play the most important role in such a process. The eNPP0f factors and their determinants could also be used to evaluate the potential of technological development for the minimization of human appropriation of NPP0 as closing global yield gaps has been suggested as one of the challenges for food security and effective agricultural production (Lobell et al., 2009). However, both those analyses are beyond the scope of this paper. In this analysis, we assess only the level of appropriation of potential productivity in terrestrial ecoregions posed by the harvesting of crops. For example, mangroves form a biome endangered by many other activities such as clearing for coastal development or shrimp aquaculture. Those biomes not suitable for agriculture have a low share of NPP0 appropriated by croplands (18% of natural potential productivity), although the appropriation of NPP and ecosystem burden can take place by other means. NPP0 have been often used as a reference for actual NPP in land use intensity studies which show that crop production especially in tropical regions is
considerably reduced in comparison to the potential productivity (Niedertscheider et al., 2016). Our results complement these findings that in terrestrial ecoregions, cropland production embodies a significant fraction of NPP0. However, the land use intensity measured by comparison of actual and potential NPP does not reveal the overall outcome of human appropriation of ecosystems. Only the tundra ecoregion can be considered relatively untouched by agriculture, followed by boreal forests (taiga). Otherwise, people have appropriated even parts of desert and xeric ecoregions. 6. Conclusions The presented analysis focuses on the extent of anthropogenic transformation of world ecoregions by cropland. We document the level of the transformation of terrestrial ecoregions by the extent of appropriation of potential net natural productivity (NPP0). Comparing terrestrial ecoregions, Temperate Grasslands, Savannas and Shrublands is the most exposed biome with 34% of NPP0 appropriated by cropland. However, in absolute terms, people appropriate most of NPP0 in Tropical and Subtropical Moist Broadleaved Forest (2.2 Gt C). The first six crops by total amount of appropriation of NPP0 (wheat, maize, rice, pumpkins for fodder, soybeans and barley) are responsible for more than half of total NPP0 appropriated by cropland world-wide (9.1 Gt C annually). Replacement of natural vegetation by croplands significantly influences the structure and functioning of habitats, contributing to biodiversity loss. Our analysis develops a novel approach to account for the transformation of ecosystems by croplands. Potential NPP footprint, introduced in this analysis, quantifies NPP of potential vegetation in terrestrial ecoregions, embodied in agricultural products that replace the original vegetation. We found that
300
et al. / Journal of Cleaner Production 150 (2017) 294e300 H. Medkova
globally, people appropriate 14% of potential net primary productivity by agriculture, but with large regional differences. In traditionally exploited regions, croplands appropriate more than 30% of potential NPP. The factors of NPP0 appropriation we developed reflect natural productivity as well as agricultural practices. Therefore, the results reported in this analysis can be used in strengthening strategies to alleviate agriculture e nature conservation conflicts in terrestrial ecoregions of the world. These findings can support strategies and policies related to food security, nature friendly agricultural practices and biodiversity conservation within planetary boundaries. Acknowledgement This work has been supported by the Czech Science Foundation via grant number 14-05292S. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jclepro.2017.03.002. References Allred, B.W., Smith, W.K., Twidwell, D., Haggerty, J.H., Running, S.W., Naugle, D.E., Fuhlendorf, S.D., 2015. Ecosystem services lost to oil and gas in North America. Science 348, 401e402. Bailey, S.A., Horner-Devine, M.C., Luck, G., Moore, L.A., Carney, K.M., Anderson, S., Betrus, C., Fleishman, E., 2004. Primary productivity and species richness: relationships among functional guilds, residency groups and vagility classes at multiple spatial scales. Ecography 27, 207e217. Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S.J., Kubiszewski, I., Farber, S., Turner, R.K., 2014. Changes in the global value of ecosystem services. Glob. Environ. Change 26, 152e158. Costanza, R., Fisher, B., Mulder, K., Liu, S., Christopher, T., 2007. Biodiversity and ecosystem services: a multi-scale empirical study of the relationship between species richness and net primary production. Ecol. Econ. 61, 478e491. DeFries, R., 2002. Past and future sensitivity of primary production to human modification of the landscape. Geophys. Res. Lett. 29, 31e36. Ellis, E.C., Goldewijk, K.K., Siebert, S., Lightman, D., Ramankutty, N., 2010. Anthropogenic transformation of the biomes, 1700 to 2000. Glob. Ecol. Biogeogr. 19, 589e606. FAO, 2015. FAOSTAT Database. Food and Agriculture Organization of the United Nations, Rome, Italy. Gibbs, H.K., Ruesch, A.S., Achard, F., Clayton, M.K., Holmgren, P., Ramankutty, N.,
Foley, J.A., 2010. Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. Proc. Natl. Acad. Sci. U. S. A. 107, 16732e16737. Haberl, H., Erb, K.H., Krausmann, F., 2014. Human appropriation of net primary production: patterns, trends, and planetary boundaries. Annu. Rev. Environ. Resour. 363e391. Haberl, H., Erb, K.H., Krausmann, F., Gaube, V., Bondeau, A., Plutzar, C., Gingrich, S., Lucht, W., Fischer-Kowalski, M., 2007. Quantifying and mapping the human appropriation of net primary production in earth's terrestrial ecosystems. Proc. Natl. Acad. Sci. U. S. A. 104, 12942e12945. Hoekstra, J.M., Boucher, T.M., Ricketts, T.H., Roberts, C., 2005. Confronting a biome crisis: global disparities of habitat loss and protection. Ecol. Lett. 8, 23e29. Imhoff, M.L., Bounoua, L., Ricketts, T., Loucks, C., Harriss, R., Lawrence, W.T., 2004. Global patterns in human consumption of net primary production. Nature 429, 870e873. Leff, B., Ramankutty, N., Foley, J.A., 2004. Geographic distribution of major crops across the world. Glob. Biogeochem. Cycles 18, 1001e1027. GB1009. Lewis, S.L., Maslin, M.A., 2015. Defining the Anthropocene. Nature 519, 171e180. Lobell, D.B., Cassman, K.G., Field, C.B., 2009. Crop yield gaps: their importance, magnitudes, and causes. Annu. Rev. Environ. Resour. 179e204. Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-being: Synthesis. Island Press, Washington, DC. Monfreda, C., Ramankutty, N., Foley, J.A., 2008. Farming the planet: 2. Geographic distribution of crop areas, yields, physiological types, and net primary production in the year 2000. Glob. Biogeochem. Cycles 22. Niedertscheider, M., Kastner, T., Fetzel, T., Haberl, H., Kroisleitner, C., Plutzar, C., Erb, K.H., 2016. Mapping and analysing cropland use intensity from a NPP perspective. Environ. Res. Lett. 11. Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V.N., Underwood, E.C., D'Amico, J.A., Itoua, I., Strand, H.E., Morrison, J.C., Loucks, C.J., Allnutt, T.F., Ricketts, T.H., Kura, Y., Lamoreux, J.F., Wettengel, W.W., Hedao, P., Kassem, K.R., 2001. Terrestrial ecoregions of the world: a new map of life on Earth. Bioscience 51, 933e938. Ramankutty, N., Evan, A.T., Monfreda, C., Foley, J.A., 2008. Farming the planet: 1. Geographic distribution of global agricultural lands in the year 2000. Glob. Biogeochem. Cycles 22. €m, J., Steffen, W., Noone, K., Persson, Å., Chapin, F.S., Lambin, E.F., Rockstro Lenton, T.M., Scheffer, M., Folke, C., Schellnhuber, H.J., Nykvist, B., De Wit, C.A., €rlin, S., Snyder, P.K., Costanza, R., Hughes, T., Van Der Leeuw, S., Rodhe, H., So Svedin, U., Falkenmark, M., Karlberg, L., Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P., Foley, J.A., 2009. A safe operating space for humanity. Nature 461, 472e475. Smith, W.K., Cleveland, C.C., Reed, S.C., Running, S.W., 2014. Agricultural conversion without external water and nutrient inputs reduces terrestrial vegetation productivity. Geophys. Res. Lett. 41, 449e455. Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., Matson, P.A., 1986. Human appropriation of the products of photosynthesis. Bioscience 36, 368e373. Watson, J.E.M., Iwamura, T., Butt, N., 2013. Mapping vulnerability and conservation adaptation strategies under climate change. Nat. Clim. Change 3, 989e994. Weinzettel, J., Hertwich, E.G., Peters, G.P., Steen-Olsen, K., Galli, A., 2013. Affluence drives the global displacement of land use. Glob. Environ. Change 23, 433e438. You, L., Wood, S., 2006. An entropy approach to spatial disaggregation of agricultural production. Agric. Syst. 90, 329e347.