- Email: [email protected]
Simulating and estimating tempo-spatial patterns in global human appropriation of net primary production (HANPP): A consumption-based approach
Ecological Indicators 23 (2012) 660–667
Contents lists available at SciVerse ScienceDirect
Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind
Simulating and estimating tempo-spatial patterns in global human appropriation of net primary production (HANPP): A consumption-based approach Ting Ma ∗ , Chenghu Zhou, Tao Pei State Key Laboratory of Resources and Environmental Information System, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
a r t i c l e
i n f o
Article history: Received 1 August 2011 Received in revised form 23 May 2012 Accepted 25 May 2012 Keywords: Terrestrial photosynthesis products Human appropriation Consumption-based HANPP Tempo-spatial pattern IPAT model
a b s t r a c t Anthropogenic alterations of biomass flows in earth’s biogeochemical cycles may profoundly affect the amount of biomass available, the level of biodiversity and the extent of carbon sequestration in global terrestrial ecosystems. Quantitative assessments of humanity’s impacts on ecosystem structures and services are therefore essential for projections of changes in terrestrial vegetation. Human appropriation of photosynthetic production (HANPP) has been extensively used as an ecological indicator for monitoring direct human interventions into terrestrial ecosystems. Here, we present the results of tempo-spatial estimations of the loss of net primary production by global terrestrial ecosystems due to human consumption-based appropriation (cHANPP, an aggregate ecological indicator for evaluating human impacts on terrestrial ecosystems due to harvesting and processing of plants for consumption) from 2000 to 2050. Our estimates are based on previously derived estimate of global biomass harvest and use for the year 2000 (Krausmann et al., 2008) through association with IPAT (Impact = Population × Affluence × Technology) model used for estimating the influence of changes in population, per capita consumption demands and technology employed during harvesting and processing of plants on biomass consumption. Our results show a distinct tendency toward increased global cHANPP by 0.17 Pg C yr−1 (P < 0.001) from 2000 to 2050 (changes resulting from land conversion are excluded), mainly resulting from an increased global population size and intensified per capita consumption of agricultural products. Long-term trends and spatial patterns in cHANPP exhibit significant variations across countries and geographical zones owing to tempo-spatial variations of both population size and consumption patterns. The proportion of potentially available photosynthetic production appropriated by human consumption, estimated at approximately ∼28% in the 2000s, is projected to increase to approximately ∼33% in the 2040s. Our results also indicate that technology may play a crucial role for alleviating the growing impact of human activities on terrestrial ecosystems and provide potential insights for sustainable development in the context of management issues and decision making. © 2012 Elsevier Ltd. All rights reserved.
1. Introduction Global climate change and the continuously escalating concentration of atmospheric carbon dioxide have been widely documented to enhance plant growth (Myneni et al., 1997; Sitch et al., 2008), particularly in northern latitudes (Myneni et al., 2001; Lucht et al., 2002), and therefore lead to a marked increase in global net primary production (NPP) (Nemani et al., 2003), the net amount of energy (mostly solar) biologically assimilated through photosynthesis. However, this might not result in an increase in the amount of NPP, actually available to all heterotrophs, because a
∗ Corresponding author. Tel.: +86 10 64889789. E-mail address: [email protected] (T. Ma). 1470-160X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ecolind.2012.05.026
loss of photosynthesis products is simultaneously caused by human appropriation through both harvest and anthropogenic changes in land cover and land use (Vitousek et al., 1986; Haberl et al., 2007; Zika and Erb, 2009). Human appropriation of net primary production (HANPP) derives from both direct and indirect consumption of terrestrial photosynthesis products through agriculture and forestry as well as the loss of biomass caused by changes in human land use (Vitousek et al., 1986). The appropriation not only reduces the food sources available to other species, but it also distinctly alters the material and energy flows in Earth’s biogeochemical cycles and within food webs (Haberl et al., 2007). HANPP is therefore an important ecological indicator of the magnitude of humanity’s impacts on terrestrial ecosystems and is useful for examining the implication of these impacts for sustainable development at both global and regional
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
scales (Haberl, 1997; Haberl et al., 2005; Imhoff and Bounoua, 2006; O’Neill et al., 2007; Bishop and Amaratunga, 2010). HANPP has received considerable attention over the past several decades in attempts to address the basic question of how much of the biosphere’s annual primary production is altered due to human intervention. Vitousek et al. (1986) conducted the first quantitative calculations of global HANPP using three estimate levels (low, intermediate and high). In these calculations, the low estimate was simply the amount of terrestrial NPP consumed directly by humans and domestic animals. The intermediate estimate calculated all the productivity of human-dominated lands entirely for socioeconomic activities. The loss of productive capacity of terrestrial ecosystems owing to land conversion was further included in the high estimate. Their results suggested that global HANPP in the early 1980s might be estimated as 7.2, 42.6 and 58.1 Pg C yr−1 (approximately 3%, 19%, and 40% of the global terrestrial NPP, respectively) for the three estimate levels, respectively. Rojstaczer et al. (2001) estimated global HANPP, in which changes in NPP resulting from land conversion were excluded, for the 1990s as 20 Pg C yr−1 (32% of the global NPP) with uncertainty of 10–55%. Imhoff et al. (2004) calculated the magnitude of global consumption-based HANPP in 1995 to be 11.5 Pg C yr−1 (approximately 20% of total NPP) using an intermediate estimate that excluded the contribution of land conversion. Haberl et al. (2007) presented a land-based estimate of 15.6 Pg C yr−1 (23.8% of the total NPP) for global HANPP around the year 2000 in which terrestrial NPP loss caused by land use change and human-induced fires were taken into account. Although previous studies differed slightly in their quantitative estimates of global HANPP, largely because of differences in their definitions of HANPP, calculation methods and data sources, the results of these previous studies commonly suggest that a considerable proportion of global NPP may be appropriated by human activities. Although spatial patterns in HANPP for a specified year have been proposed (Imhoff et al., 2004; Krausmann et al., 2009), the long-term trends in global HANPP in relation to changes in world population size and spatial distribution, alterations in human demands on agricultural and forest products and the levels of technology employed in agricultural and forestry sectors, however, are not well addressed. Investigating shifts in the tempo-spatial patterns of global HANPP can help us to further assess anthropogenic effects on terrestrial ecosystems in terms of global carbon cycling, world food security, the provision of ecosystem services and global sustainability concerns (Haberl et al., 2007). The primary objective of this study is to quantitatively investigate tempo-spatial variations in global consumption-based HANPP (cHANPP), which is defined as the loss of terrestrial NPP due to direct human consumption of biomass for food, feed and fiber as well as biomass loss during harvest. Based on the previous calculation of global biomass harvest and use by Krausmann et al. (2008), we simulated the tempo-spatial changes in global cHANPP for the period from 2000 to 2050 using the IPAT model (Impact = Population × Affluence × Technology) (Dietz and Rosa, 1994, 1997) in combination with three datasets: (1) the time series data of country-level agricultural and forest product consumption derived from the Food and Agriculture Organization (FAO, 2010), (2) country-level population size derived from the FAO (2010) and (3) potential terrestrial NPP derived from dynamic global vegetation models (Cramer et al., 2001). We then performed comparative analyses regarding tempo-spatial distribution patterns in global cHANPP: (1) the impacts of world population size, human demands and technology on cHANPP dynamics, (2) the differences in cHANPP trends between countries and latitudinal zones and (3) the changes in the spatial pattern of grid-based cHANPP.
661
2. Materials and methods 2.1. Calculation of cHANPP From an ecological perspective, HANPP can been defined as the difference between the potential net primary production (NPP0 ), the solar energy incorporated by the prevailing vegetation in the absence of human activities, and the NPP of the actual vegetation remaining in ecosystems after harvest (Haberl et al., 2007; O’Neill et al., 2007; Erb et al., 2009; Krausmann et al., 2009). Three different components are usually taken into account in the calculation of HANPP: (1) organic materials used directly by humans and domestic animals, (2) biomass destroyed and left on site during harvest, and (3) the loss of productive capacity of ecosystems due to human-induced land use changes, land degradation and humaninduced fires. The results on estimation of HANPP may therefore vary depending upon the definition and scope used. In this study, we defined HANPP as the sum of components (1) and (2). Our estimates for long-term tempo-spatial trends in HANPP from 2000 to 2050 started with FAO country level consumption data in association with gridded population density data. We therefore use the term “cHANPP” to emphasize that we pay attention to the assessment of the “upstream” biomass losses associated with the consumption of biomass-based products (Haberl et al., 2009) while we do not include changes in NPP resulting from land conversion. Krausmann et al. (2008) provided a comprehensive estimation of global patterns of socioeconomic biomass flows around the year 2000 derived from FAO country-level consumption datasets. Based on their quantitative approach to estimating global HANPP, we calculated cHANPP as the sum of three distinct components: food-related cHANPP, wood-related cHANPP and unused extraction (cHANPPunused ): cHANPP = cHANPPfood + cHANPPwood + cHANPPunused
(1)
where food-related cHANPP was defined as terrestrial NPP losses due to harvested crops, used crop residues and grazed biomass (in relation to animal food consumption). The amount of NPP lost due to wood (for building and fuel), paper and fiber consumption was treated as wood-related cHANPP. Unused extraction of NPP was defined as indirect human appropriation of terrestrial ecosystem production through unused residuals from both crop and wood harvests and belowground biomass of harvested crops and felled trees (Krausmann et al., 2008). Aquatic biomass consumption and the influences of transportation of photosynthetic products through regional and global trades were not taken into account in our estimates of tempo-spatial patterns in cHANPP because of the lack of data. The long-term trends in country-level cHANPP were estimated using the IPAT model (Impact = Population × Affluence × Technology) (Dietz and Rosa, 1994, 1997) in association with quantitative estimates of cHANPP in the year 2000 for 175 countries provided by Krausmann et al. (2008) as: cHANPP(Y ) = PoM(Y ) × TeM(Y ) × [FoM(Y ) × cHANPPfood (2000) + WoM(Y ) × cHANPPwood (2000) + (˛FoM(Y ) + ˇFiM(Y )) × cHANPPunused (2000)]
(2)
where PoM, FoM, WoM and TeM are normalized multipliers of country-level population size, food-related consumption, wood-related consumption and technology employed during the harvesting and processing of biomass in the year Y with respect to the corresponding values in the year 2000. ˛ and ˇ are proportionality constants of food-related consumption and wood-related consumption in the year Y, respectively.
662
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
In addition, to project production and consumption balance for global terrestrial net primary production, we used the average output of six dynamic global vegetation models with inputs of the HadCM2 climate scenario and the IPCC IS92a scenario of rising CO2 to estimate global terrestrial NPP0 for the period from 2000 to 2050 (Cramer et al., 2001). 2.2. Estimates of the food-related, wood-related and population multipliers Krausmann et al. (2008) estimated global patterns of socioeconomics biomass for the year 2000 mainly using FAO country-level datasets of agricultural and forest products consumed in 1998–2002 for 175 countries. In their calculations, various harvest and efficiency factors were applied to terrestrial biomass consumption in different types of crops, forests and feed demands across different regions. In our study, for simplicity, we used time series data of per capita food supply from 1999 to 2007 (FAO, 2010) to estimate annual fluctuations of food-related multiplier (FoM) for each country. Long-term trends in FoM from 2008 to 2050 were simulated using power law regression analysis. To estimate annual variations in wood-related multiplier (FiM), firstly, we assembled a country-level dataset of annual forestry production including roundwood, sawnwood, wood fuel and wood pulp (FAO, 2010) from 1999 to 2009. We then calculated per capita consumption of forestry products and used power law regression analysis to estimate the trajectories of per capita demand for wood-relation products during 2010–2050 for every country. We finally normalized per capita food supply and per capita consumption of forestry products to the three-year average (1999–2001) (in order to reduce estimate errors due to data uncertainty) to derive country-level estimates of food-related multiplier and wood-related multiplier in Eq. (2) for a given year Y: FoM(Y ) =
FiM(Y ) =
food supply in the year Y average food supply of 1999–2001
(3)
wood-related consumption in the year Y average wood-related consumption of 1999–2001 (4)
In our calculations, increased food-related multiplier and woodrelated multiplier may imply escalating per capita consumption demands for agricultural and forestry products, respectively. Moreover, in order to reduce error propagation caused by power law fitting of country-level consumption data, the maximum values of food-related multiplier and wood-related multiplier are limited to 2. Time series data from PopSTAT (FAO, 2010), which provides estimated and projected country-level population data for the period from 1999 to 2050, were used to calculate temporal variations in country-level population multiplier (PoM) in Eq. (2) as: PoM(Y ) =
population in the year Y average population of 1999–2001
(5)
2.3. Estimates of the technology multiplier IPAT model (see Eq. (2)) provides a simple mathematical approach for measuring the relationship between humanity’s impact on the environment and technological innovation (Chertow, 2000). Enhanced production efficiency during the harvesting and processing of terrestrial biomass and productivity of human-dominated ecosystems due to improvement in technology used could reduce the requirements of photosynthetic production (Wirsenius, 2003). In IPAT model, the technology term
Fig. 1. (a) Long-term variations of global cHANPP, food-related cHANPP and woodrelated cHANPP from 2000 to 2050. Gray-shaded area shows estimated trend in global cHANPP without consideration of the technology factor in the IPAT model. (b) Estimated annual fluctuations (black) and trends (red) in proportion between global cHANPP and terrestrial NPP0 from 2000 to 2050. (c) Statistical distributions of increased (red) and decreased (blue) country-level multipliers between 2000 and 2050 in the IPAT model. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
generally incorporates various driving factors beyond population size and affluence that reflect the impact of human activity on the environment and resources, including social organization, culture and institutions (Dietz and Rosa, 1994, 1997). For this study, technology factor therefore should be an operational measure for comprehensively quantifying the changes in production efficiency of human-dominated ecosystems and harvest factors during biomass harvesting and processing over time. In our IPAT model, we considered the term of technology as the ecological impact of per unit photosynthetic production for human consumption. Thus, enhanced technology used during biomass harvesting and processing may imply reduced human appropriation of photosynthetic production for per unit consumption of agricultural and forest products. An indicator for measuring ecological impact of per unit biomass production therefore is required to quantify the term of technology in our IPAT model. Energy intensity, defined as the total primary energy consumption per dollar of gross domestic product using purchasing power parities, has been extensively used in various IPAT models for representing the environment impact per unit production (Chertow, 2000; York et al., 2003). Here, we have chosen time series data of country-level energy intensity from 1994 to 2006 derived from the US Energy Information Administration
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
663
Fig. 2. Comparisons of estimated annual variations in (a) cHANPP, (b) food-related cHANPP, (c) wood-related cHANPP, (d) unused cHANPP, and (e) population size between developed, developing and least developed countries from 2000 to 2050. (f) Comparisons of statistical distribution of increased (red) and decreased (cyan) multipliers in different countries. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
(2006) as the proxy for measuring changes in the ecological impact of per unit human consumption of food-related and wood-related products on photosynthetic biomass of terrestrial ecosystems. The time series data of energy intensity from 1994 to 2006 was used to statistically predict the long-term trends in countrylevel energy intensity for the period of 2007–2050 according to the power law. We then transformed all country-level time series of energy intensity data into the normalized technology multiplier TeM in Eq. (2) in the year Y with respect to the mean of 1999–2001: TeM(Y ) =
energy intensity in year Y average energy intensity of 1999–2001
(6)
provides spatially explicit human population densities in a common geo-referenced framework, was used as basic map for the spatial estimates of the local human population for all grid cells. Because spatial distributions of the human population are available for only few time points (2000, 2005, 2010 and 2015 for our study) in GPWv3, the country-level population derived from PopSTAT (FAO, 2010) in a given year was proportionally allocated into all contained grids according to the grid-level value of population density in the nearest year available in GPWv3. Thus, the global patterns of cHANPP were finally mapped as spatial aggregations of global land grids in which grid-level cHANPP were calculated by multiplying population numbers in each grid cell with countryspecific per capita cHANPP for the period of 2000–2050.
2.4. Mapping grid-based cHANPP 2.5. Estimates of trends in cHANPP To map the spatial patterns of global cHANPP, we first created a gridded global land dataset at 0.5◦ × 0.5◦ geographical resolution using the WGS84 spatial reference system. We then positioned all grids on a world country map using overlay operations with geographical information system (GIS) software to obtain quantitative information regarding country-level cHANPP. Furthermore, the spatial distribution of population density within each grid cell was required for the calculation of cHANPP at grid-level. The Gridded Population of the World map (version 3, GPWv3, produced by the Center for International Earth Science Information Network of the Earth Institute at Columbia University, 2005), which
Based on the results of both country-level and gridded estimates of cHANPP, we assessed tempo-spatial patterns in global cHANPP and their relationships to changes in human population size, food-related consumption, wood-related consumptions and technology. Comparative analyses of trends in cHANPP across different countries and latitudinal zones were performed to further delineate the tempo-spatial variability of cHANPP. Linear regression analysis was applied for trend estimates of cHANPP and statistical significance of trend analysis was tested with two-tailed t-tests.
664
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
3. Results and discussion 3.1. Temporal trends in global cHANPP Our results show that global cHANPP, in which biomass loss due to human-induced fires and land conversions was excluded, is estimated to significantly increase from 16.20 Pg C in 2000 (Krausmann et al., 2008) to 24.27 Pg C in 2050 at a linear rate of 0.17 Pg C yr−1 (P < 0.001) (Fig. 1(a)). Meanwhile, seventy percent of the increase in global cHANPP is likely due to food-related (including crop and livestock) consumption (0.12 Pg C yr−1 , P < 0.001), and only 6% of the increase is derived from wood-related consumption (0.01 Pg C yr−1 , P < 0.001). Unused extraction of terrestrial biomass contributes 24% of the increase in global cHANPP. The strong positive trend in global cHANPP results primarily from both world population growth (from 6.1 billion in 2000 to a projected 9.1 billion in 2050) and the escalation of per capita demands for terrestrial biomass, particularly for agricultural products. As shown in Fig. 1(c), 95% and 83% of countries show increased population size and per capita food consumption from 2000 to 2050, respectively. Consequently, increases in food-related cHANPP are likely resulted from growths of both human population and per capita demand for agricultural products. Increased woodrelated cHANPP, however, is mainly driven by increased population because positive trends in wood-related multiplier from 2000 to 2050 are found in only 30% of countries. Moreover, improved technology (i.e. decreased technology multiplier, occurring in 89% of countries) is likely to reduce the growth of human appropriation of terrestrial biomass because the fact is that cHANNP is projected to reach 28.06 Pg C in 2050 without consideration of technology multiplier in our IPAT model. The remaining terrestrial NPP after consumption-related harvesting and processing is projected to grow by 0.09 Pg C yr−1 (P < 0.001) between 2000 and 2050, which is noticeably less than the synchronous climate-driven increase in potential global NPP by 0.26 Pg C yr−1 (P < 0.001) due to the concurrently rising human appropriation. Consequently, the proportion of human consumption-induced reduction of global potential terrestrial NPP is projected to increase considerably from ∼28% in 2000s to ∼33% in 2040s (Fig. 1(b)).
Fig. 3. Comparisons of estimated annual variations in the country-average of (a) food-related multiplier and (b) wood-related multiplier between developed, developing and least developed countries from 2000 to 2050. (c) Trends in proportions of “saved” NPP by improving technology used (2000–2050) for different countries.
3.2. Comparisons of trends in cHANPP between countries Country-level cHANPP is not determined only by the population size, it is also related to the level of economic development, which usually influences the needs of per capita consumption for agricultural and forest products and the magnitude of technology employed in socioeconomic activities. To compare temporal trends in cHANPP between different countries, we grouped all countries into three categories: developed countries (consisting of 43 countries), developing countries (93 countries) and least-developed countries (37 countries) based on data from the United Nations Statistics Division (2011). Figs. 2 and 3 illustrate conspicuous differences in the estimated trends in cHANPP and its impact factors between the three categories of countries for the period from 2000 to 2050. As shown in Fig. 2(a), both least developed countries and developing countries commonly show a marked growth rate of cHANPP by 0.05 Pg C yr−1 (P < 0.001) and 0.10 Pg C yr−1 (P < 0.001), respectively. These rates are significantly greater than that estimated for developed countries, 0.02 Pg C yr−1 (P < 0.001). Population growth and a prominent increase in per capita demand for food-related products are likely two key factors in the rapidly increasing cHANPP over the next few decades in the least-developed countries (Figs. 2(e) and 3(a)). Furthermore, a continuously rising energy intensity multiplier (i.e. aggravated negative influences on terrestrial ecosystems; see Fig. 2(f)) in 19%
of least-developed countries will likely exacerbate the humaninduced losses of terrestrial NPP in these countries. For developing countries, the significant trend toward increased cHANPP is mainly derived from both human population growth (Fig. 2(e)) and a rapidly increasing per capita consumption of food-related products (Fig. 3(a)). Escalating per capita consumption requirement of food-related products is a major driving force behind the increasing cHANPP in developed countries (Fig. 3(a)), in which population size (Fig. 2(e)) and per capita demand for wood-related production (Fig. 3(b)) show only slight increase and distinct decrease from 2000 to 2050, respectively. Moreover, improving technology in developed countries will likely produces conspicuous reductions in cHANPP (Fig. 2(f)); approximately 20% of cHANPP is projected to be “saved” in 2050 due to notable decrease in technology multiplier in developed countries (Fig. 3(c)). 3.3. Comparisons of trends in cHANPP between latitudinal zones Climate-driven increase in NPP0 at a mean rate of 0.10 Pg C yr−1 (P < 0.001) is predicted for the tropical zone (30S–20N) by six dynamic global vegetation models (Cramer et al., 2001) (Fig. 4(a)). However, the proportion of between cHANPP and NPP0 is projected to markedly rise from 20% in the 2000s to 31% in the 2040s for
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
665
Fig. 4. Estimated annual variations of NPP0 (a), cHANPP (b), and the proportion between cHANPP and NPP0 (c) from 2000 to 2050 for different latitudinal zones. (d) Area proportions of three latitudinal zones.
this region (Fig. 4(c)) owing to a significant increase in cHANPP at a rate of 0.11 Pg C yr−1 (P < 0.001) (Fig. 4(b)). This result may imply an intensifying human impact on tropical ecosystems in the coming decades. Negative human impacts on terrestrial NPP are likely to be more conspicuous in the temperate zone (20N–40N) because in this region, the demands of human consumption account for nearly 78% of the potential terrestrial NPP (see Fig. 4(c)). In this region, both cHANPP and NPP0 show distinct increases by 0.06 Pg C yr−1 (P < 0.001) and 0.05 Pg C yr−1 (P < 0.001), respectively. In the northern zone (>40N), however, a markedly diminishing proportion between human demand and ecosystem production is projected from 23% in the 2000s to 20% in the 2040s owing to increased difference between changes in NPP0 (by 0.09 Pg C yr−1 with P < 0.001) and cHANPP (by 0.007 Pg C yr−1 with P < 0.001) from 2000 to 2050. 3.4. Changes in spatial patterns of global cHANPP The spatial patterns of cHANPP are primarily determined by the spatial density of the human population and regional variations in both per capita consumption of terrestrial photosynthetic products and technology employed during harvesting and processing of terrestrial plant. To obtain an estimate of changes in global patterns of cHANPP, we mapped the spatial distributions and long term trends in grid-based cHANPP from 2000 to 2050. As shown in Fig. 5, significant trends toward increases in cHANPP most likely occur in South Asia, Central Africa and some regions of South America in the coming decades. Moreover, trends in grid-based cHANPP show noticeable spatial variations owing to spatially uneven development of influencing factors. For instance, although the human population is expected to continuously rise in eastern China and southern Africa in the next several decades, cHANPP will likely decline due to improvements in the technology used in these regions for the period from 2000 to 2050. Declining cHANPP is also predicted for Europe, Russia and Japan, mainly because of improving technology. Increasing HANPP in India, North America and South America will likely result from two key factors: human population growth and enhanced per capita consumption. In most African countries, increased cHANPP will likely result not only from increases in both human population size and per
capita consumption but also from intensified human influences on ecosystems due to the low level of technology employed in agricultural and forestry production systems. 3.5. Errors and uncertainty of estimates In this study, the accuracy of temporal modeling of population size, per capita consumption and energy intensity, which depends on the quality of the original datasets and the simulation methods, may profoundly influence the estimates of trends in global cHANPP. Although our results show that most of the country-level time series data of per-capita consumption and energy intensity follow a power law with a statistical significance level of 0.05, it is still difficult to accurately predict the future trajectories of both the human demand for photosynthetic products and technology. Moreover, the extent to which country-level energy intensity parallels the effects of technology used in the IPAT model affects our estimates of the magnitude of human intervention in terrestrial ecosystems. Energy intensity is only operational measures of technology used in the IPAT model, but not a direct indicator which can accurately quantify the relationship between changes in technology employed during harvesting and processing of terrestrial biomass and efficiency factors in calculation of cHANPP. A direct indicator for the technology term in the IPAT model therefore is required to obtain more accurate estimates for long-term dynamics in global cHANPP due to changes in efficiency of production and harvest factors. The accuracy of multi-decadal projected trends in global cHANPP might be limited due to only one decade of time series data were used in our calculations. Spatial errors and uncertainty in cHANPP estimation could result from human migration and issues concerning the transportation of photosynthetic products, which are not explicitly taken into account in our estimates due to the limited availability of data sources. An improvement in data availability of historical records regarding human consumption of biomass production and development of a more accurate model for predicting temporal trends in the parameters used in the IPAT model, particularly for the term of technology, are essential for furthering our understanding of changes in tempo-spatial patterns of regional and global cHANPP.
666
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
Fig. 5. (a) and (b) Spatial patterns in global cHANPP for the year 2000 and the year 2050, respectively. (c) and (d) Long-term trends and P-value in global grid-based cHANPP from 2000 to 2050.
4. Conclusions Our results suggest that the net loss of the photosynthetic products of terrestrial ecosystems caused by human appropriation could continuously rise over the next several decades, mainly due to marked increases in both the global population and per capita consumption of forest and agricultural products. Consequently, the actual increase in the terrestrial NPP remaining in ecosystems and
available for other species is projected to be distinctly less than the climate-driven increase in potential terrestrial NPP. This result suggests that noticeable human-induced alterations of global carbon flow dynamics, influenced by global warming, might occur. Improving technology may partially alleviate the negative impacts of human appropriation of NPP on terrestrial ecosystems, particularly in developed and developing countries. In the least developed countries, however, poor technology likely aggravates the human
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
influence on terrestrial ecosystems as both the human population and per capita consumption are projected to rise. Rapid growth in both human population size and personal demand for photosynthetic products are primary factors in the increase in cHANPP in developing countries. In developed countries, however, increasing cHANPP results mainly from an increase in the per capita demand for agricultural products. The spatial patterns of global cHANPP and its trends vary between different countries and geographical zones. An increase in human impacts on terrestrial NPP and a growing difference between human demand and ecosystem supply most likely occur in regions with high population density. In these regions, external inputs of photosynthetic products are required to satisfy the needs of the local human population and sustain the development of ecosystems. The positive net balance between the potential supply of photosynthetic products from terrestrial ecosystems and the human demand for these photosynthetic products could be reduced after 2050, when the net productivity of terrestrial ecosystems is projected to decline due to increased ecosystem respiration induced by a warming climate (Cramer et al., 2001; Sitch et al., 2008). Improving technology, particularly in both developing countries and the least developed countries, reducing the consumption of forest products by recycling and developing substitutes, and offsetting local deficits of photosynthetic products through transportation are potential means for reducing human-induced negative impacts on the environment and for sustaining the stability and development of terrestrial ecosystems in a changing world. Acknowledgments This study was supported by National Key Technology R&D Program (No. 2011BAH06B03 and No. 2011BAH24B10). Additional supports were provided by Natural Science Foundation of China (No. 40830529 to CHZ) and Chinese Academy of Sciences (No. KZCX2-YW-QN303 to TP). We would like to gratefully thank the anonymous reviewers for their insightful and helpful comments to improve the manuscript. References Bishop, J.D.K., Amaratunga, G.A.J., 2010. Quantifying the limits of HANPP and carbon emissions with prolong total species well-being. Environ. Dev. Sustain. 12, 213–231. Center for International Earth Science Information Network (CIESIN), Columbia University; and Centro Internacional de Agricultura Tropical (CIAT), 2005. Gridded Population of the World Version 3 (GPWv3): Population Grids. Socioeconomic Data and Applications Center (SEDAC), Columbia University, Palisades, NY, Available from: http://sedac.ciesin.columbia.edu/gpw. Chertow, M.R., 2000. The IPAT equation and its variants. J. Ind. Ecol. 4 (4), 13–29.
667
Cramer, W., et al., 2001. Global response of terrestrial ecosystem structure and function to CO2 and climate change results from six dynamic global vegetation models. Global Change Biol. 7, 357–373. Dietz, T., Rosa, E.A., 1994. Rethink the environmental impacts of population, affluence and technology. Hum. Ecol. 1, 277–300. Dietz, T., Rosa, E.A., 1997. Effects of population and affluence on CO2 emissions. Proc. Natl. Acad. Sci. U.S.A. 94, 175–179. Energy Information Administration (EIA), 2006. International Energy Annual 2006. U.S. Department of Energy. Erb, K.H., et al., 2009. Analyzing the global human appropriation of net primary production-processes, trajectories, implications. An introduction. Ecol. Econ. 69, 250–259. Food and Agriculture Organization, 2010. FAOSTAT 2010 Database. UNFAO, Rome. Haberl, H., 1997. Human appropriation of net primary production as an environmental indicator: implications for sustainable development. Ambio 26 (3), 143–146. Haberl, H., Putzar, C., Erb, K.H., Gaube, V., Pollheimer, M., Schulz, N.B., 2005. Human appropriation of net primary production as determinant of avifauna diversity in Austria. Agric. Ecosyst. Environ. 110, 119–131. Haberl, H., et al., 2007. Quantifying and mapping the human appropriation of net primary production in earth’s terrestrial ecosystems. Proc. Natl. Acad. Sci. U.S.A. 104 (31), 12942–12947. Haberl, H., et al., 2009. Using embodies HANPP to analyze teleconnections in the global land system: conceptual considerations. Geografisk Tidsskrift-Danish J. Geogr. 109 (2), 119–130. Imhoff, M., Bounoua, L., Ricketts, T., Loucks, C., Harriss, R., Lawrence, W.T., 2004. Global patterns in human consumption of net primary production. Nature 429, 870–873. Imhoff, M., Bounoua, L., 2006. Exploring global patterns of net primary production carbon supply and demand using satellite observations and statistical data. J. Geophys. Res. 111 (D22S12), http://dx.doi.org/10.1029/2006JD007377. Krausmann, F., Erb, K.H., Gingrich, S., Lauk, C., Haberl, H., 2008. Global patterns of socioeconomic biomass flows in the year 2000: a comprehensive assessment of supply, consumption and constraints. Ecol. Econ. 65, 471–487. Krausmann, F., Haberl, H., Erb, K.H., Wiesinger, M., Gaube, V., Gingrich, S., 2009. What determines geographical patterns of the global human appropriation of net primary production? J. Land Use Sci. 4 (1–2), 15–23. Lucht, W., et al., 2002. Climatic control of the high-latitude vegetation greening trend and Pinatubo effect. Science 296, 1687–1689. Myneni, R.B., Keeling, C.D., Tucker, C.J., Asrar, G., Nemani, R.R., 1997. Increased plant growth in the northern high latitudes from 1981–1991. Nature 386, 698–702. Myneni, R.B., et al., 2001. A large carbon sink in the woody biomass of Northern forests. Proc. Natl. Acad. Sci. U.S.A. 98, 14784–14789. Nemani, R.R., et al., 2003. Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science 300, 1560–1563. O’Neill, D.W., Tyedmers, P.H., Beazley, K.F., 2007. Human appropriation of net primary production (HANPP) in Nova Scotia, Canada. Reg. Environ. Change 7, 1–14. Rojstaczer, S., Sterling, S.M., Moore, N.J., 2001. Human appropriation of photosynthesis products. Science 294, 2549–2552. Sitch, S., et al., 2008. Evaluation of the terrestrial carbon cycle, future plant geography and climate-carbon cycle feedbacks using five Dynamic Global Vegetation Models. Global Change Biol. 14, 2015–2039. United Nations Statistics Division, 2011. Composition of macro geographical (continental) regions, geographical sub-regions, and selected economic and other groupings.. Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., Matson, P.A., 1986. Human appropriation of the products of photosynthesis. Bioscience 36 (6), 368–373. Wirsenius, S., 2003. Efficiencies and biomass appropriation of food commodities on global and regional levels. Agric. Syst. 77 (3), 219–255. York, R., Rosa, E.A., Dietz, T., 2003. STIRPAT, IPAT and ImPACT: analytic tools for unpacking the driving forces of environmental impacts. Ecol. Econ. 46, 351–365. Zika, M., Erb, K.H., 2009. The global loss of net primary production resulting from human-induced soil degradation in drylands. Ecol. Econ. 69 (2), 310–318.
Contents lists available at SciVerse ScienceDirect
Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind
Simulating and estimating tempo-spatial patterns in global human appropriation of net primary production (HANPP): A consumption-based approach Ting Ma ∗ , Chenghu Zhou, Tao Pei State Key Laboratory of Resources and Environmental Information System, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
a r t i c l e
i n f o
Article history: Received 1 August 2011 Received in revised form 23 May 2012 Accepted 25 May 2012 Keywords: Terrestrial photosynthesis products Human appropriation Consumption-based HANPP Tempo-spatial pattern IPAT model
a b s t r a c t Anthropogenic alterations of biomass flows in earth’s biogeochemical cycles may profoundly affect the amount of biomass available, the level of biodiversity and the extent of carbon sequestration in global terrestrial ecosystems. Quantitative assessments of humanity’s impacts on ecosystem structures and services are therefore essential for projections of changes in terrestrial vegetation. Human appropriation of photosynthetic production (HANPP) has been extensively used as an ecological indicator for monitoring direct human interventions into terrestrial ecosystems. Here, we present the results of tempo-spatial estimations of the loss of net primary production by global terrestrial ecosystems due to human consumption-based appropriation (cHANPP, an aggregate ecological indicator for evaluating human impacts on terrestrial ecosystems due to harvesting and processing of plants for consumption) from 2000 to 2050. Our estimates are based on previously derived estimate of global biomass harvest and use for the year 2000 (Krausmann et al., 2008) through association with IPAT (Impact = Population × Affluence × Technology) model used for estimating the influence of changes in population, per capita consumption demands and technology employed during harvesting and processing of plants on biomass consumption. Our results show a distinct tendency toward increased global cHANPP by 0.17 Pg C yr−1 (P < 0.001) from 2000 to 2050 (changes resulting from land conversion are excluded), mainly resulting from an increased global population size and intensified per capita consumption of agricultural products. Long-term trends and spatial patterns in cHANPP exhibit significant variations across countries and geographical zones owing to tempo-spatial variations of both population size and consumption patterns. The proportion of potentially available photosynthetic production appropriated by human consumption, estimated at approximately ∼28% in the 2000s, is projected to increase to approximately ∼33% in the 2040s. Our results also indicate that technology may play a crucial role for alleviating the growing impact of human activities on terrestrial ecosystems and provide potential insights for sustainable development in the context of management issues and decision making. © 2012 Elsevier Ltd. All rights reserved.
1. Introduction Global climate change and the continuously escalating concentration of atmospheric carbon dioxide have been widely documented to enhance plant growth (Myneni et al., 1997; Sitch et al., 2008), particularly in northern latitudes (Myneni et al., 2001; Lucht et al., 2002), and therefore lead to a marked increase in global net primary production (NPP) (Nemani et al., 2003), the net amount of energy (mostly solar) biologically assimilated through photosynthesis. However, this might not result in an increase in the amount of NPP, actually available to all heterotrophs, because a
∗ Corresponding author. Tel.: +86 10 64889789. E-mail address: [email protected] (T. Ma). 1470-160X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ecolind.2012.05.026
loss of photosynthesis products is simultaneously caused by human appropriation through both harvest and anthropogenic changes in land cover and land use (Vitousek et al., 1986; Haberl et al., 2007; Zika and Erb, 2009). Human appropriation of net primary production (HANPP) derives from both direct and indirect consumption of terrestrial photosynthesis products through agriculture and forestry as well as the loss of biomass caused by changes in human land use (Vitousek et al., 1986). The appropriation not only reduces the food sources available to other species, but it also distinctly alters the material and energy flows in Earth’s biogeochemical cycles and within food webs (Haberl et al., 2007). HANPP is therefore an important ecological indicator of the magnitude of humanity’s impacts on terrestrial ecosystems and is useful for examining the implication of these impacts for sustainable development at both global and regional
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
scales (Haberl, 1997; Haberl et al., 2005; Imhoff and Bounoua, 2006; O’Neill et al., 2007; Bishop and Amaratunga, 2010). HANPP has received considerable attention over the past several decades in attempts to address the basic question of how much of the biosphere’s annual primary production is altered due to human intervention. Vitousek et al. (1986) conducted the first quantitative calculations of global HANPP using three estimate levels (low, intermediate and high). In these calculations, the low estimate was simply the amount of terrestrial NPP consumed directly by humans and domestic animals. The intermediate estimate calculated all the productivity of human-dominated lands entirely for socioeconomic activities. The loss of productive capacity of terrestrial ecosystems owing to land conversion was further included in the high estimate. Their results suggested that global HANPP in the early 1980s might be estimated as 7.2, 42.6 and 58.1 Pg C yr−1 (approximately 3%, 19%, and 40% of the global terrestrial NPP, respectively) for the three estimate levels, respectively. Rojstaczer et al. (2001) estimated global HANPP, in which changes in NPP resulting from land conversion were excluded, for the 1990s as 20 Pg C yr−1 (32% of the global NPP) with uncertainty of 10–55%. Imhoff et al. (2004) calculated the magnitude of global consumption-based HANPP in 1995 to be 11.5 Pg C yr−1 (approximately 20% of total NPP) using an intermediate estimate that excluded the contribution of land conversion. Haberl et al. (2007) presented a land-based estimate of 15.6 Pg C yr−1 (23.8% of the total NPP) for global HANPP around the year 2000 in which terrestrial NPP loss caused by land use change and human-induced fires were taken into account. Although previous studies differed slightly in their quantitative estimates of global HANPP, largely because of differences in their definitions of HANPP, calculation methods and data sources, the results of these previous studies commonly suggest that a considerable proportion of global NPP may be appropriated by human activities. Although spatial patterns in HANPP for a specified year have been proposed (Imhoff et al., 2004; Krausmann et al., 2009), the long-term trends in global HANPP in relation to changes in world population size and spatial distribution, alterations in human demands on agricultural and forest products and the levels of technology employed in agricultural and forestry sectors, however, are not well addressed. Investigating shifts in the tempo-spatial patterns of global HANPP can help us to further assess anthropogenic effects on terrestrial ecosystems in terms of global carbon cycling, world food security, the provision of ecosystem services and global sustainability concerns (Haberl et al., 2007). The primary objective of this study is to quantitatively investigate tempo-spatial variations in global consumption-based HANPP (cHANPP), which is defined as the loss of terrestrial NPP due to direct human consumption of biomass for food, feed and fiber as well as biomass loss during harvest. Based on the previous calculation of global biomass harvest and use by Krausmann et al. (2008), we simulated the tempo-spatial changes in global cHANPP for the period from 2000 to 2050 using the IPAT model (Impact = Population × Affluence × Technology) (Dietz and Rosa, 1994, 1997) in combination with three datasets: (1) the time series data of country-level agricultural and forest product consumption derived from the Food and Agriculture Organization (FAO, 2010), (2) country-level population size derived from the FAO (2010) and (3) potential terrestrial NPP derived from dynamic global vegetation models (Cramer et al., 2001). We then performed comparative analyses regarding tempo-spatial distribution patterns in global cHANPP: (1) the impacts of world population size, human demands and technology on cHANPP dynamics, (2) the differences in cHANPP trends between countries and latitudinal zones and (3) the changes in the spatial pattern of grid-based cHANPP.
661
2. Materials and methods 2.1. Calculation of cHANPP From an ecological perspective, HANPP can been defined as the difference between the potential net primary production (NPP0 ), the solar energy incorporated by the prevailing vegetation in the absence of human activities, and the NPP of the actual vegetation remaining in ecosystems after harvest (Haberl et al., 2007; O’Neill et al., 2007; Erb et al., 2009; Krausmann et al., 2009). Three different components are usually taken into account in the calculation of HANPP: (1) organic materials used directly by humans and domestic animals, (2) biomass destroyed and left on site during harvest, and (3) the loss of productive capacity of ecosystems due to human-induced land use changes, land degradation and humaninduced fires. The results on estimation of HANPP may therefore vary depending upon the definition and scope used. In this study, we defined HANPP as the sum of components (1) and (2). Our estimates for long-term tempo-spatial trends in HANPP from 2000 to 2050 started with FAO country level consumption data in association with gridded population density data. We therefore use the term “cHANPP” to emphasize that we pay attention to the assessment of the “upstream” biomass losses associated with the consumption of biomass-based products (Haberl et al., 2009) while we do not include changes in NPP resulting from land conversion. Krausmann et al. (2008) provided a comprehensive estimation of global patterns of socioeconomic biomass flows around the year 2000 derived from FAO country-level consumption datasets. Based on their quantitative approach to estimating global HANPP, we calculated cHANPP as the sum of three distinct components: food-related cHANPP, wood-related cHANPP and unused extraction (cHANPPunused ): cHANPP = cHANPPfood + cHANPPwood + cHANPPunused
(1)
where food-related cHANPP was defined as terrestrial NPP losses due to harvested crops, used crop residues and grazed biomass (in relation to animal food consumption). The amount of NPP lost due to wood (for building and fuel), paper and fiber consumption was treated as wood-related cHANPP. Unused extraction of NPP was defined as indirect human appropriation of terrestrial ecosystem production through unused residuals from both crop and wood harvests and belowground biomass of harvested crops and felled trees (Krausmann et al., 2008). Aquatic biomass consumption and the influences of transportation of photosynthetic products through regional and global trades were not taken into account in our estimates of tempo-spatial patterns in cHANPP because of the lack of data. The long-term trends in country-level cHANPP were estimated using the IPAT model (Impact = Population × Affluence × Technology) (Dietz and Rosa, 1994, 1997) in association with quantitative estimates of cHANPP in the year 2000 for 175 countries provided by Krausmann et al. (2008) as: cHANPP(Y ) = PoM(Y ) × TeM(Y ) × [FoM(Y ) × cHANPPfood (2000) + WoM(Y ) × cHANPPwood (2000) + (˛FoM(Y ) + ˇFiM(Y )) × cHANPPunused (2000)]
(2)
where PoM, FoM, WoM and TeM are normalized multipliers of country-level population size, food-related consumption, wood-related consumption and technology employed during the harvesting and processing of biomass in the year Y with respect to the corresponding values in the year 2000. ˛ and ˇ are proportionality constants of food-related consumption and wood-related consumption in the year Y, respectively.
662
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
In addition, to project production and consumption balance for global terrestrial net primary production, we used the average output of six dynamic global vegetation models with inputs of the HadCM2 climate scenario and the IPCC IS92a scenario of rising CO2 to estimate global terrestrial NPP0 for the period from 2000 to 2050 (Cramer et al., 2001). 2.2. Estimates of the food-related, wood-related and population multipliers Krausmann et al. (2008) estimated global patterns of socioeconomics biomass for the year 2000 mainly using FAO country-level datasets of agricultural and forest products consumed in 1998–2002 for 175 countries. In their calculations, various harvest and efficiency factors were applied to terrestrial biomass consumption in different types of crops, forests and feed demands across different regions. In our study, for simplicity, we used time series data of per capita food supply from 1999 to 2007 (FAO, 2010) to estimate annual fluctuations of food-related multiplier (FoM) for each country. Long-term trends in FoM from 2008 to 2050 were simulated using power law regression analysis. To estimate annual variations in wood-related multiplier (FiM), firstly, we assembled a country-level dataset of annual forestry production including roundwood, sawnwood, wood fuel and wood pulp (FAO, 2010) from 1999 to 2009. We then calculated per capita consumption of forestry products and used power law regression analysis to estimate the trajectories of per capita demand for wood-relation products during 2010–2050 for every country. We finally normalized per capita food supply and per capita consumption of forestry products to the three-year average (1999–2001) (in order to reduce estimate errors due to data uncertainty) to derive country-level estimates of food-related multiplier and wood-related multiplier in Eq. (2) for a given year Y: FoM(Y ) =
FiM(Y ) =
food supply in the year Y average food supply of 1999–2001
(3)
wood-related consumption in the year Y average wood-related consumption of 1999–2001 (4)
In our calculations, increased food-related multiplier and woodrelated multiplier may imply escalating per capita consumption demands for agricultural and forestry products, respectively. Moreover, in order to reduce error propagation caused by power law fitting of country-level consumption data, the maximum values of food-related multiplier and wood-related multiplier are limited to 2. Time series data from PopSTAT (FAO, 2010), which provides estimated and projected country-level population data for the period from 1999 to 2050, were used to calculate temporal variations in country-level population multiplier (PoM) in Eq. (2) as: PoM(Y ) =
population in the year Y average population of 1999–2001
(5)
2.3. Estimates of the technology multiplier IPAT model (see Eq. (2)) provides a simple mathematical approach for measuring the relationship between humanity’s impact on the environment and technological innovation (Chertow, 2000). Enhanced production efficiency during the harvesting and processing of terrestrial biomass and productivity of human-dominated ecosystems due to improvement in technology used could reduce the requirements of photosynthetic production (Wirsenius, 2003). In IPAT model, the technology term
Fig. 1. (a) Long-term variations of global cHANPP, food-related cHANPP and woodrelated cHANPP from 2000 to 2050. Gray-shaded area shows estimated trend in global cHANPP without consideration of the technology factor in the IPAT model. (b) Estimated annual fluctuations (black) and trends (red) in proportion between global cHANPP and terrestrial NPP0 from 2000 to 2050. (c) Statistical distributions of increased (red) and decreased (blue) country-level multipliers between 2000 and 2050 in the IPAT model. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
generally incorporates various driving factors beyond population size and affluence that reflect the impact of human activity on the environment and resources, including social organization, culture and institutions (Dietz and Rosa, 1994, 1997). For this study, technology factor therefore should be an operational measure for comprehensively quantifying the changes in production efficiency of human-dominated ecosystems and harvest factors during biomass harvesting and processing over time. In our IPAT model, we considered the term of technology as the ecological impact of per unit photosynthetic production for human consumption. Thus, enhanced technology used during biomass harvesting and processing may imply reduced human appropriation of photosynthetic production for per unit consumption of agricultural and forest products. An indicator for measuring ecological impact of per unit biomass production therefore is required to quantify the term of technology in our IPAT model. Energy intensity, defined as the total primary energy consumption per dollar of gross domestic product using purchasing power parities, has been extensively used in various IPAT models for representing the environment impact per unit production (Chertow, 2000; York et al., 2003). Here, we have chosen time series data of country-level energy intensity from 1994 to 2006 derived from the US Energy Information Administration
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
663
Fig. 2. Comparisons of estimated annual variations in (a) cHANPP, (b) food-related cHANPP, (c) wood-related cHANPP, (d) unused cHANPP, and (e) population size between developed, developing and least developed countries from 2000 to 2050. (f) Comparisons of statistical distribution of increased (red) and decreased (cyan) multipliers in different countries. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
(2006) as the proxy for measuring changes in the ecological impact of per unit human consumption of food-related and wood-related products on photosynthetic biomass of terrestrial ecosystems. The time series data of energy intensity from 1994 to 2006 was used to statistically predict the long-term trends in countrylevel energy intensity for the period of 2007–2050 according to the power law. We then transformed all country-level time series of energy intensity data into the normalized technology multiplier TeM in Eq. (2) in the year Y with respect to the mean of 1999–2001: TeM(Y ) =
energy intensity in year Y average energy intensity of 1999–2001
(6)
provides spatially explicit human population densities in a common geo-referenced framework, was used as basic map for the spatial estimates of the local human population for all grid cells. Because spatial distributions of the human population are available for only few time points (2000, 2005, 2010 and 2015 for our study) in GPWv3, the country-level population derived from PopSTAT (FAO, 2010) in a given year was proportionally allocated into all contained grids according to the grid-level value of population density in the nearest year available in GPWv3. Thus, the global patterns of cHANPP were finally mapped as spatial aggregations of global land grids in which grid-level cHANPP were calculated by multiplying population numbers in each grid cell with countryspecific per capita cHANPP for the period of 2000–2050.
2.4. Mapping grid-based cHANPP 2.5. Estimates of trends in cHANPP To map the spatial patterns of global cHANPP, we first created a gridded global land dataset at 0.5◦ × 0.5◦ geographical resolution using the WGS84 spatial reference system. We then positioned all grids on a world country map using overlay operations with geographical information system (GIS) software to obtain quantitative information regarding country-level cHANPP. Furthermore, the spatial distribution of population density within each grid cell was required for the calculation of cHANPP at grid-level. The Gridded Population of the World map (version 3, GPWv3, produced by the Center for International Earth Science Information Network of the Earth Institute at Columbia University, 2005), which
Based on the results of both country-level and gridded estimates of cHANPP, we assessed tempo-spatial patterns in global cHANPP and their relationships to changes in human population size, food-related consumption, wood-related consumptions and technology. Comparative analyses of trends in cHANPP across different countries and latitudinal zones were performed to further delineate the tempo-spatial variability of cHANPP. Linear regression analysis was applied for trend estimates of cHANPP and statistical significance of trend analysis was tested with two-tailed t-tests.
664
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
3. Results and discussion 3.1. Temporal trends in global cHANPP Our results show that global cHANPP, in which biomass loss due to human-induced fires and land conversions was excluded, is estimated to significantly increase from 16.20 Pg C in 2000 (Krausmann et al., 2008) to 24.27 Pg C in 2050 at a linear rate of 0.17 Pg C yr−1 (P < 0.001) (Fig. 1(a)). Meanwhile, seventy percent of the increase in global cHANPP is likely due to food-related (including crop and livestock) consumption (0.12 Pg C yr−1 , P < 0.001), and only 6% of the increase is derived from wood-related consumption (0.01 Pg C yr−1 , P < 0.001). Unused extraction of terrestrial biomass contributes 24% of the increase in global cHANPP. The strong positive trend in global cHANPP results primarily from both world population growth (from 6.1 billion in 2000 to a projected 9.1 billion in 2050) and the escalation of per capita demands for terrestrial biomass, particularly for agricultural products. As shown in Fig. 1(c), 95% and 83% of countries show increased population size and per capita food consumption from 2000 to 2050, respectively. Consequently, increases in food-related cHANPP are likely resulted from growths of both human population and per capita demand for agricultural products. Increased woodrelated cHANPP, however, is mainly driven by increased population because positive trends in wood-related multiplier from 2000 to 2050 are found in only 30% of countries. Moreover, improved technology (i.e. decreased technology multiplier, occurring in 89% of countries) is likely to reduce the growth of human appropriation of terrestrial biomass because the fact is that cHANNP is projected to reach 28.06 Pg C in 2050 without consideration of technology multiplier in our IPAT model. The remaining terrestrial NPP after consumption-related harvesting and processing is projected to grow by 0.09 Pg C yr−1 (P < 0.001) between 2000 and 2050, which is noticeably less than the synchronous climate-driven increase in potential global NPP by 0.26 Pg C yr−1 (P < 0.001) due to the concurrently rising human appropriation. Consequently, the proportion of human consumption-induced reduction of global potential terrestrial NPP is projected to increase considerably from ∼28% in 2000s to ∼33% in 2040s (Fig. 1(b)).
Fig. 3. Comparisons of estimated annual variations in the country-average of (a) food-related multiplier and (b) wood-related multiplier between developed, developing and least developed countries from 2000 to 2050. (c) Trends in proportions of “saved” NPP by improving technology used (2000–2050) for different countries.
3.2. Comparisons of trends in cHANPP between countries Country-level cHANPP is not determined only by the population size, it is also related to the level of economic development, which usually influences the needs of per capita consumption for agricultural and forest products and the magnitude of technology employed in socioeconomic activities. To compare temporal trends in cHANPP between different countries, we grouped all countries into three categories: developed countries (consisting of 43 countries), developing countries (93 countries) and least-developed countries (37 countries) based on data from the United Nations Statistics Division (2011). Figs. 2 and 3 illustrate conspicuous differences in the estimated trends in cHANPP and its impact factors between the three categories of countries for the period from 2000 to 2050. As shown in Fig. 2(a), both least developed countries and developing countries commonly show a marked growth rate of cHANPP by 0.05 Pg C yr−1 (P < 0.001) and 0.10 Pg C yr−1 (P < 0.001), respectively. These rates are significantly greater than that estimated for developed countries, 0.02 Pg C yr−1 (P < 0.001). Population growth and a prominent increase in per capita demand for food-related products are likely two key factors in the rapidly increasing cHANPP over the next few decades in the least-developed countries (Figs. 2(e) and 3(a)). Furthermore, a continuously rising energy intensity multiplier (i.e. aggravated negative influences on terrestrial ecosystems; see Fig. 2(f)) in 19%
of least-developed countries will likely exacerbate the humaninduced losses of terrestrial NPP in these countries. For developing countries, the significant trend toward increased cHANPP is mainly derived from both human population growth (Fig. 2(e)) and a rapidly increasing per capita consumption of food-related products (Fig. 3(a)). Escalating per capita consumption requirement of food-related products is a major driving force behind the increasing cHANPP in developed countries (Fig. 3(a)), in which population size (Fig. 2(e)) and per capita demand for wood-related production (Fig. 3(b)) show only slight increase and distinct decrease from 2000 to 2050, respectively. Moreover, improving technology in developed countries will likely produces conspicuous reductions in cHANPP (Fig. 2(f)); approximately 20% of cHANPP is projected to be “saved” in 2050 due to notable decrease in technology multiplier in developed countries (Fig. 3(c)). 3.3. Comparisons of trends in cHANPP between latitudinal zones Climate-driven increase in NPP0 at a mean rate of 0.10 Pg C yr−1 (P < 0.001) is predicted for the tropical zone (30S–20N) by six dynamic global vegetation models (Cramer et al., 2001) (Fig. 4(a)). However, the proportion of between cHANPP and NPP0 is projected to markedly rise from 20% in the 2000s to 31% in the 2040s for
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
665
Fig. 4. Estimated annual variations of NPP0 (a), cHANPP (b), and the proportion between cHANPP and NPP0 (c) from 2000 to 2050 for different latitudinal zones. (d) Area proportions of three latitudinal zones.
this region (Fig. 4(c)) owing to a significant increase in cHANPP at a rate of 0.11 Pg C yr−1 (P < 0.001) (Fig. 4(b)). This result may imply an intensifying human impact on tropical ecosystems in the coming decades. Negative human impacts on terrestrial NPP are likely to be more conspicuous in the temperate zone (20N–40N) because in this region, the demands of human consumption account for nearly 78% of the potential terrestrial NPP (see Fig. 4(c)). In this region, both cHANPP and NPP0 show distinct increases by 0.06 Pg C yr−1 (P < 0.001) and 0.05 Pg C yr−1 (P < 0.001), respectively. In the northern zone (>40N), however, a markedly diminishing proportion between human demand and ecosystem production is projected from 23% in the 2000s to 20% in the 2040s owing to increased difference between changes in NPP0 (by 0.09 Pg C yr−1 with P < 0.001) and cHANPP (by 0.007 Pg C yr−1 with P < 0.001) from 2000 to 2050. 3.4. Changes in spatial patterns of global cHANPP The spatial patterns of cHANPP are primarily determined by the spatial density of the human population and regional variations in both per capita consumption of terrestrial photosynthetic products and technology employed during harvesting and processing of terrestrial plant. To obtain an estimate of changes in global patterns of cHANPP, we mapped the spatial distributions and long term trends in grid-based cHANPP from 2000 to 2050. As shown in Fig. 5, significant trends toward increases in cHANPP most likely occur in South Asia, Central Africa and some regions of South America in the coming decades. Moreover, trends in grid-based cHANPP show noticeable spatial variations owing to spatially uneven development of influencing factors. For instance, although the human population is expected to continuously rise in eastern China and southern Africa in the next several decades, cHANPP will likely decline due to improvements in the technology used in these regions for the period from 2000 to 2050. Declining cHANPP is also predicted for Europe, Russia and Japan, mainly because of improving technology. Increasing HANPP in India, North America and South America will likely result from two key factors: human population growth and enhanced per capita consumption. In most African countries, increased cHANPP will likely result not only from increases in both human population size and per
capita consumption but also from intensified human influences on ecosystems due to the low level of technology employed in agricultural and forestry production systems. 3.5. Errors and uncertainty of estimates In this study, the accuracy of temporal modeling of population size, per capita consumption and energy intensity, which depends on the quality of the original datasets and the simulation methods, may profoundly influence the estimates of trends in global cHANPP. Although our results show that most of the country-level time series data of per-capita consumption and energy intensity follow a power law with a statistical significance level of 0.05, it is still difficult to accurately predict the future trajectories of both the human demand for photosynthetic products and technology. Moreover, the extent to which country-level energy intensity parallels the effects of technology used in the IPAT model affects our estimates of the magnitude of human intervention in terrestrial ecosystems. Energy intensity is only operational measures of technology used in the IPAT model, but not a direct indicator which can accurately quantify the relationship between changes in technology employed during harvesting and processing of terrestrial biomass and efficiency factors in calculation of cHANPP. A direct indicator for the technology term in the IPAT model therefore is required to obtain more accurate estimates for long-term dynamics in global cHANPP due to changes in efficiency of production and harvest factors. The accuracy of multi-decadal projected trends in global cHANPP might be limited due to only one decade of time series data were used in our calculations. Spatial errors and uncertainty in cHANPP estimation could result from human migration and issues concerning the transportation of photosynthetic products, which are not explicitly taken into account in our estimates due to the limited availability of data sources. An improvement in data availability of historical records regarding human consumption of biomass production and development of a more accurate model for predicting temporal trends in the parameters used in the IPAT model, particularly for the term of technology, are essential for furthering our understanding of changes in tempo-spatial patterns of regional and global cHANPP.
666
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
Fig. 5. (a) and (b) Spatial patterns in global cHANPP for the year 2000 and the year 2050, respectively. (c) and (d) Long-term trends and P-value in global grid-based cHANPP from 2000 to 2050.
4. Conclusions Our results suggest that the net loss of the photosynthetic products of terrestrial ecosystems caused by human appropriation could continuously rise over the next several decades, mainly due to marked increases in both the global population and per capita consumption of forest and agricultural products. Consequently, the actual increase in the terrestrial NPP remaining in ecosystems and
available for other species is projected to be distinctly less than the climate-driven increase in potential terrestrial NPP. This result suggests that noticeable human-induced alterations of global carbon flow dynamics, influenced by global warming, might occur. Improving technology may partially alleviate the negative impacts of human appropriation of NPP on terrestrial ecosystems, particularly in developed and developing countries. In the least developed countries, however, poor technology likely aggravates the human
T. Ma et al. / Ecological Indicators 23 (2012) 660–667
influence on terrestrial ecosystems as both the human population and per capita consumption are projected to rise. Rapid growth in both human population size and personal demand for photosynthetic products are primary factors in the increase in cHANPP in developing countries. In developed countries, however, increasing cHANPP results mainly from an increase in the per capita demand for agricultural products. The spatial patterns of global cHANPP and its trends vary between different countries and geographical zones. An increase in human impacts on terrestrial NPP and a growing difference between human demand and ecosystem supply most likely occur in regions with high population density. In these regions, external inputs of photosynthetic products are required to satisfy the needs of the local human population and sustain the development of ecosystems. The positive net balance between the potential supply of photosynthetic products from terrestrial ecosystems and the human demand for these photosynthetic products could be reduced after 2050, when the net productivity of terrestrial ecosystems is projected to decline due to increased ecosystem respiration induced by a warming climate (Cramer et al., 2001; Sitch et al., 2008). Improving technology, particularly in both developing countries and the least developed countries, reducing the consumption of forest products by recycling and developing substitutes, and offsetting local deficits of photosynthetic products through transportation are potential means for reducing human-induced negative impacts on the environment and for sustaining the stability and development of terrestrial ecosystems in a changing world. Acknowledgments This study was supported by National Key Technology R&D Program (No. 2011BAH06B03 and No. 2011BAH24B10). Additional supports were provided by Natural Science Foundation of China (No. 40830529 to CHZ) and Chinese Academy of Sciences (No. KZCX2-YW-QN303 to TP). We would like to gratefully thank the anonymous reviewers for their insightful and helpful comments to improve the manuscript. References Bishop, J.D.K., Amaratunga, G.A.J., 2010. Quantifying the limits of HANPP and carbon emissions with prolong total species well-being. Environ. Dev. Sustain. 12, 213–231. Center for International Earth Science Information Network (CIESIN), Columbia University; and Centro Internacional de Agricultura Tropical (CIAT), 2005. Gridded Population of the World Version 3 (GPWv3): Population Grids. Socioeconomic Data and Applications Center (SEDAC), Columbia University, Palisades, NY, Available from: http://sedac.ciesin.columbia.edu/gpw. Chertow, M.R., 2000. The IPAT equation and its variants. J. Ind. Ecol. 4 (4), 13–29.
667
Cramer, W., et al., 2001. Global response of terrestrial ecosystem structure and function to CO2 and climate change results from six dynamic global vegetation models. Global Change Biol. 7, 357–373. Dietz, T., Rosa, E.A., 1994. Rethink the environmental impacts of population, affluence and technology. Hum. Ecol. 1, 277–300. Dietz, T., Rosa, E.A., 1997. Effects of population and affluence on CO2 emissions. Proc. Natl. Acad. Sci. U.S.A. 94, 175–179. Energy Information Administration (EIA), 2006. International Energy Annual 2006. U.S. Department of Energy. Erb, K.H., et al., 2009. Analyzing the global human appropriation of net primary production-processes, trajectories, implications. An introduction. Ecol. Econ. 69, 250–259. Food and Agriculture Organization, 2010. FAOSTAT 2010 Database. UNFAO, Rome. Haberl, H., 1997. Human appropriation of net primary production as an environmental indicator: implications for sustainable development. Ambio 26 (3), 143–146. Haberl, H., Putzar, C., Erb, K.H., Gaube, V., Pollheimer, M., Schulz, N.B., 2005. Human appropriation of net primary production as determinant of avifauna diversity in Austria. Agric. Ecosyst. Environ. 110, 119–131. Haberl, H., et al., 2007. Quantifying and mapping the human appropriation of net primary production in earth’s terrestrial ecosystems. Proc. Natl. Acad. Sci. U.S.A. 104 (31), 12942–12947. Haberl, H., et al., 2009. Using embodies HANPP to analyze teleconnections in the global land system: conceptual considerations. Geografisk Tidsskrift-Danish J. Geogr. 109 (2), 119–130. Imhoff, M., Bounoua, L., Ricketts, T., Loucks, C., Harriss, R., Lawrence, W.T., 2004. Global patterns in human consumption of net primary production. Nature 429, 870–873. Imhoff, M., Bounoua, L., 2006. Exploring global patterns of net primary production carbon supply and demand using satellite observations and statistical data. J. Geophys. Res. 111 (D22S12), http://dx.doi.org/10.1029/2006JD007377. Krausmann, F., Erb, K.H., Gingrich, S., Lauk, C., Haberl, H., 2008. Global patterns of socioeconomic biomass flows in the year 2000: a comprehensive assessment of supply, consumption and constraints. Ecol. Econ. 65, 471–487. Krausmann, F., Haberl, H., Erb, K.H., Wiesinger, M., Gaube, V., Gingrich, S., 2009. What determines geographical patterns of the global human appropriation of net primary production? J. Land Use Sci. 4 (1–2), 15–23. Lucht, W., et al., 2002. Climatic control of the high-latitude vegetation greening trend and Pinatubo effect. Science 296, 1687–1689. Myneni, R.B., Keeling, C.D., Tucker, C.J., Asrar, G., Nemani, R.R., 1997. Increased plant growth in the northern high latitudes from 1981–1991. Nature 386, 698–702. Myneni, R.B., et al., 2001. A large carbon sink in the woody biomass of Northern forests. Proc. Natl. Acad. Sci. U.S.A. 98, 14784–14789. Nemani, R.R., et al., 2003. Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science 300, 1560–1563. O’Neill, D.W., Tyedmers, P.H., Beazley, K.F., 2007. Human appropriation of net primary production (HANPP) in Nova Scotia, Canada. Reg. Environ. Change 7, 1–14. Rojstaczer, S., Sterling, S.M., Moore, N.J., 2001. Human appropriation of photosynthesis products. Science 294, 2549–2552. Sitch, S., et al., 2008. Evaluation of the terrestrial carbon cycle, future plant geography and climate-carbon cycle feedbacks using five Dynamic Global Vegetation Models. Global Change Biol. 14, 2015–2039. United Nations Statistics Division, 2011. Composition of macro geographical (continental) regions, geographical sub-regions, and selected economic and other groupings.