Reviews on Impact Assessments of Land-Use Change on Key Ecosystem Services

Fig. 1.1
Diagram to trace the impacts of LUC and climate variations on the human well-being via altering the ecosystem provisioning services

Exploration on the Functions of LUC on Ecosystem Provisioning Services

The ecosystems can provide a variety of direct and indirect services to humans and other living organisms, and those services can be affected by climate variations and human activities, especially human-induced LUC (Bangash et al. 2013). Some studies have shown that LUC and climate variations are among the greatest global environmental pressures resulting from anthropogenic activities, which significantly influence the provision of crucial ecosystem services, such as carbon sequestration, water flow regulation, and food and fiber production, at a variety of scales (Reyers et al. 2009). In the researches of the relationship between LUC and ecosystem services, the analyses of the impacts of LUC on ecosystem services were usually conducted from the aspect of land-use quantity and structure changes. Some studies have indicated that the diversification of land use would help to improve ecosystem services (Swift et al. 2004). In addition, combined effects of LUC and climate variations may change ecosystem services, especially the food provision and water yield of agro-ecosystem, and forest and/or grassland ecosystems (Sun et al. 2006).

Influences of LUC on the Aagro-Ecosystem Services

As a kind of specific complex manual–natural ecosystem, agro-ecosystem has not only efficient and direct production function, but also the function of environmental services, tourism services, and aesthetic services (Swinton et al. 2007). Agriculture is a dominant form of land management globally, and agro-ecosystem covers nearly 40 % of the terrestrial surface of the earth (Power 2010). Agro-ecosystem is faced with severe challenges under the context of global warming and the intensive human-induced LUC.

More and more current studies have shown that LUC especially rapid urbanization has already directly or indirectly affected the food provisioning services of agro-ecosystem (Jaradat and Boody 2012). The quality and quantity changes of agricultural land have potential effects on provision services of agro-ecosystem. However, some studies have shown that agricultural land use has degraded the soil, water, net primary productivity (NPP), and the biological assets in agro-ecosystem to such an extent that the restoration of natural capital and rehabilitation of ecosystem services are needed through changes in land use and management (Foley et al. 2011). A primary reason for this degradation is the failure of agricultural commodity markets to internalize environmental costs associated with land use and management decisions.

The LUC especially the excessive reclamation of cropland and the intensive agriculture land use exerts potential effects on biodiversity conservation of agro-ecosystem, which will influence the stability of provisioning services in agro-ecosystem. Some studies have indicated that the rapid expansion and intensification of row crop production have resulted in the loss of habitat and spatial heterogeneity in agro-ecosystem, which affects the ecosystem provisioning services (Gavier-Pizarro et al. 2012). Land-use conversion from natural lands to croplands, grazing lands, and urban areas has been increased over time, resulted in reduced or modified biodiversity, altered functional processes, and diminished provision of ecosystem goods and services to society globally (DIAz et al. 2007). Some studies also indicated that the intensive agricultural development could change land use, which can further affect regional ecosystem services (Dale and Polasky 2007).

Influences of LUC on the Ecosystem Services from Forest and/or Grassland

Forest and grassland ecosystems are indispensable constituent parts of terrestrial ecosystems which play important roles in global climate variations. Climate variations affect the water yield of forest and/or grassland ecosystems via its direct influence on precipitation and evaporation process of atmosphere hydrologic cycle. LUC, including converting grassland or shrublands to plantations, afforestation, and reforestation, is gaining attention globally and will alter many ecosystem processes, including water yield of forest and grassland ecosystems (Farley et al. 2005; Nosetto et al. 2005). Changes in the extent and composition of forest, grassland, wetland, and other ecosystems have large impacts on the biophysical conditions, which further affect the provision of ecosystem services and biodiversity conservation. The LUC influences the water yield of ecosystems through changing the transpiration, interception, and evaporation, all of which tend to increase when grassland or scrubland is replaced with forests. Transpiration rates are influenced by changes in rooting characteristics, leaf area, stomata response, plant surface albedo, and turbulence (Vertessy et al. 2001).

Much progress has been made to understand the effect of LUC on water yield of forest ecosystems during the past century all over the world, and the results of which generally indicated that LUC has both positive and negative effects on water yield. For example, clear-cutting forests in the USA may result in the increase in annual water yield (Ice and Stednick 2004). The vegetation restoration will have positive effects on watershed health by reducing soil erosion and nonpoint source pollution, enhancing terrestrial and aquatic habitat, and increasing ecosystem carbon sequestration (Sun et al. 2006). Some studies also indicated that vegetation changes, particularly those involving transitions between forests and grasslands dominated covers, often modify evaporative water losses as a result of plant-mediated shifts in moisture access and demand. And massive afforestation of native grassland has strong yet poorly quantified effects on the hydrological cycle (Nosetto et al. 2005). Since forests with well-developed root systems cost plentiful groundwater and soil water, it can save plenty of water to convert forest into short seasonal crops. To plant abundant pasture instead of forest in catchment areas is becoming a widely used method to increase water yield.

Quantitative Identification of the Impacts of LUC on Ecosystem Services

Many ecologists and natural scientists study ecosystem processes to understand ecosystem services across different landscapes via quantifying ecosystem services (Raudsepp-Hearne et al. 2010). Quantification and valuation of services, if linked with payments or incentives, can enhance policies and regulations that properly reward decisions that yield public benefits. It is well known that ecosystems are essential to the existence of humans, while ecosystem services are typically not priced correctly at their value because of absence of markets for ecosystem goods and services and inadequate or nonexistent information about the value of goods and services. There are many studies evaluating the impacts of LUC on ecosystem services in both developed and developing countries (Su et al. 2012). To sum up the quantitative researches so far, we found that there are two kinds of methods to value the ecosystem services, namely observations based on remote sensing and GIS technology and modeling approaches.

Enhanced Observation and Valuation Approaches of Ecosystem Services

The historical ecosystem service value could be reflected by remote sensing along with a GIS-based model (GEOMOD) since LUC is of the upper most driving forces of regional ecosystems and has huge impacts on ecosystem service value. Remote sensing provides reliable area-wide data for quantifying and mapping ecosystem services at comparatively low costs, and with the fast, frequent, and continuous observations for monitoring. GEOMOD model is a kind of method and technique to allocate LUC spatially and to evaluate its impact on ecosystems simply and transparently.

The selection of indicators to be used in the analyses of potential impacts of LUC is the main challenge after obtaining the remote sensing data. The valuation of different ecosystem services and the spatial–temporal monitoring of their respective changes can provide useful indicators of the potential impacts of LUC. Generally speaking, as the ecosystem service values were different for each land-use category (Estoque and Murayama 2012), Costanza et al. (1997) first attempted to estimate the ecosystem service value coefficients (Costanza et al. 1997), and then, many researches use the same approach in order to quantify and map the ecosystem service values at global or regional scales (Turner et al. 2007). Although there are many potential conceptual and empirical problems and limitations to estimate the ecosystem service values (Nelson et al. 2009), the magnitude of the estimated ecosystem service value changes in the LUC is substantial. Thus, it may still be possible to draw general inferences about the effect of the perceived LUC on the estimated ecosystem service values.

Improved Models for Assessing Ecosystem Services

More and more ecosystem service values were assessed by models around the world in recent decades, and it seems to have become a trend to assess various ecosystem services with models. To avoid the weakness of common assessment models used before, some improved approaches or models were formed to assess the ecosystem service values, including the Millennium Ecosystem Assessment (MEA) approach (Chopra et al. 2005), Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) (Tallis and Polasky 2009) model, and the UK National Ecosystem Assessment (UK NEA) (Assessment 2011).

The MEA was set up in 2000, and it is the first time to provide a comprehensive picture of past, present, and possible future trends in ecosystem services and their values and propose corresponding measures. Many researchers have assessed the ecosystem services at national scale with framework of MEA. They have valued the agricultural ecosystem services, forest ecosystem services, and grassland ecosystem services in China (Xie et al. 2006). For instance, 17 ecosystem services in 18 categories of grassland ecosystem in China have been assessed (Xiao et al. 2003). Exploring the researches between LUC and ecosystem services, we found that most studies analyzed the impacts by analyzing the quantity and structure change of land use. It is generally acknowledged that the provision of ecosystem services depends on biophysical conditions and changes over space and time due to human-induced LUC. Spatial patterns of LUC can be linked to large regions and provide direct measures of human activities (Riitters et al. 2002).

The InVEST model has been widely used in valuating ecosystem service values (Polasky et al. 2011). The model uses maps and tabular data of land use and land management in conjunction with environmental information, such as soil, topography, and climate, to generate spatially explicit predictions of the ecosystem services. InVEST model estimates the provision and value of ecosystem services under alternative land-use scenarios. Economic information about demand for ecosystem services can be combined with biophysical supply to generate predictive maps of services use and values (Xie et al. 2006; Daily et al. 2009). InVEST model also analyzes the impacts of land use and land management on species habitat provision and quality. Thus, the model provides a powerful tool for quantifying and valuing multiple ecosystem services and assessing the impacts of LUC. By varying land use or land management and evaluating the corresponding output with InVEST, we can provide useful information to managers and policy-makers to weigh the trade-offs in ecosystem services, biodiversity conservation, and other land-use objectives.

The UK NEA is the first analysis of the UK’s natural environment in terms of the benefits it provides to society and the nation’s continuing prosperity. It has been a wide-ranging, multi-stakeholder, cross-disciplinary process, designed to provide a comprehensive picture of past, present, and possible future trends in ecosystem services and their values; it is underpinned by the best available evidence and the most up-to-date conceptual thinking and analytical tools, which can be applied to assess the ecosystem service values aimed to describe the changes of key drivers that affect the UK’s ecosystems, including changes in land use and climate. The UK NEA distinguished between the ecosystem processes and intermediate ecosystem services and the final ecosystem services that directly deliver welfare gains and/or losses to people. This distinction is important to avoid double counting in the valuation of ecosystem services (Fisher et al. 2008). As the researches go on, the UK NEA would have a broad application prospects to assess the impacts of human activities on ecosystem services.

At any rate, ecosystem services play an important role in maintaining the balance of global ecosystems and improving human living environment. Quantifying and mapping those ecosystem services is necessary to periodically determine the response of ecosystem services to global change, such as LUC and climate variations. Those approaches and models, though being widely used to quantify and map the ecosystem service values around the world, still have potential to be improved in order to get more accurate assessment results. In addition, so far there is still no assessment system and method that has been commonly approved by researchers. Therefore, it is still a hot issue to study the assessment of ecosystem service values to clarify the relationship between LUC and ecosystem services.

Major Indicators Identifying Influences of Ecosystem Provisioning Services on Human Well-Being

Ecosystem services are essential for the maintenance of human well-being, and the links between ecosystem services and human well-being are complex, diverse, and complicated to assess properly with the consideration of different spatial and temporal scales (Pereira et al. 2005). Healthy ecosystems provide services that are the foundation for human well-being including the provision of resources for basic survival, such as clean air, water, and genetic resources for medicines, along with the provision of raw materials for industry and agriculture (Daily et al. 1997). Thus, the degradation and loss of ecosystem services have negative effect on human well-being. On the one hand, the degradation and loss of ecosystem provisioning services will increase the inputs in production to recover reduced outputs in the ecosystems. On the other hand, the degradation and loss of provisioning ecosystem services will increase the human health risk.

Increased Inputs with Reduced Outputs in Production

Ecosystems are changed by the LUC and climate variations, all of which may reduce or increase the supply of ecosystem services temporarily or permanently. Some evidences have shown that the climate variations and human activities especially LUC have changed agricultural and natural ecosystems more rapidly and extensively over the last 50 years (Reid et al. 2005).The MA reported that 15 of the world’s 24 ecosystem services are in decline, which have affected human well-being and threaten the survival of other species. The declining ability of the earth’s systems to meet the needs of a growing population and sustain the life support systems of the planet is a very urgent and serious issue. Due to the impacts of climate variation and LUC on agro-ecosystem services, the outputs humans obtained in agricultural production will decrease, including the food production and water yield. Therefore, humans would input more production factors such as fertilizers and pesticides in the process of food production and invest more in search of more water resources to sustain the continual ecosystem provision of human well-being.

The development of novel chemicals and new technologies during the last century has supported the modern agricultural revolution, resulting in an increase in food production and harvest rates (Galis et al. 2013). Some studies have indicated that the application of fertilizers (such as the nitrogen application) and pesticides in some regions increased rapidly to meet the demands for greater food production needs (Nelson et al. 2006). The increasing use of these chemicals and technologies removed the constraint of nutrient limitation for crop growth, as well as competitive pests and weeds, resulting in increased outputs. However, on the other side, these increased outputs are at the cost of the reducing ecosystem services, especially the provisioning services. Thus, human well-being would be weakened. With the ecosystem services changing, governments have supplied subsidies and grants to adjust personal and business behavior.

Augmented Health Risk Induced by the Irrational Land Uses

Ecosystem services can support the fundamental need of human well-being in a variety of ways. The changes of ecosystem services would both directly and indirectly put the humans’ health at risk through the insufficient provision of food and freshwater, inorganic chemicals and persistent organic chemical pollutants in food and water, and infectious disease caused by ecosystem service loss. Besides, the indirect health risk was caused by the irrational land uses.

The insufficient accesses to the ecosystem provisioning services of food production and water yield are particularly important factors leading to the health risks in human well-being. Some studies have indicated that a lack of access to the ecosystem provisioning services of food causes far more than physical harm, and it may put thousands of millions of people in mental and physical potential risks by reducing intelligent and physical growth, in some cases from the moment of human conception (Reid et al. 2005). Undernutrition was recently assessed as an underlying cause of death each year worldwide, which is particularly common in sub-Saharan Africa and south Asia, especially in India (Caulfield et al. 2004).Vegetation especially the grass and forest are important for the interception of water. Some studies have shown that about 1 billion people were affected by land degradation caused by soil erosion, waterlogging, or salinity of irrigated land.

Humans are also at risk due to inorganic chemicals and persistent organic chemical pollutants in food and water, and the infectious disease caused by ecosystem service loss. Some studies indicated human actions, for example, releasing toxic chemicals into the environment, or using pesticide and chemical fertilizer, will pollute the water and food, which can have adverse effects on various organ systems (Aktar et al. 2009). Some evidence indicated that some chemicals from pesticide and chemical fertilizer have increased the microbial contamination of drinking water, which has led to the infectious diseases accounting for approximately 6 % of all deaths globally. The pattern and extent of change in incidence of particular infectious disease depends on the particular ecosystems affected, such as type of LUC. Some studies showed that climate change and some LUC, such as deforestation, might alter infectious disease patterns (Patz et al. 2003; Sehgal 2010). There have been lots of studies to investigate the influence of the increased income and the health risk of human well-being, which will be helpful to profoundly understand the influence mechanism and extent of the degradation of ecosystem services on human well-being.

Concluding Remarks

Based on current researches about the effects of LUC and climate variations on ecosystem services, we mainly focus on ecosystem provisioning services and the influence of the changes in provisioning services on human well-being. Firstly, we explored the researches on identification and quantification of the impacts of LUC on ecosystem service values and latter examined how the impact on ecosystem provisioning services affects human well-being through analyses of the increased inputs and the reduced outputs of agricultural production and the augmented health risk of humans.

So far, there are still some researches to be done to uncover the impacts of LUC and climate variations on human well-being via ecosystem provisioning services. First of all, the current researches are focused on the ecosystem service values, but the mechanisms through which LUC and climate variations influence ecosystem services are still not well understood. However, this is of great significance to the sustainable development of human well-being. Secondly, there would be uncertainty involved when the remote sensing data were used for quantifying and mapping ecosystem services. Therefore, validating the reliability of the results obtained by using remote sensing in quantifying and mapping ecosystem services needs to be done in further researches. As to the impacts of ecosystem service changes on human well-being, more researches on quantification of these impacts need to be done to make the research more comprehensive.

It has been shown by research practice that the assessment of ecosystem services was useful to the setting of strategies and policies, with potentially far-reaching influence on human activities. Thus, further researches are needed to focus on the following three issues. First, there is a need to formulate a set of thorough and normative method to assess ecosystem service values and improve the accuracy of assessment results. Second, an in-depth process-based analysis of the relationship between human activities and ecosystem service function is needed. Third, there is an urgent need to promote the application of ecosystem service values in various aspects of production, livelihood, and government decision making and eventually serve for human well-being.

Impacts of Land-Use Change on Climate Regulation Services


Land-use change (LUC) and climate change are two major factors that result in the changes of ecosystem services (Schröter et al. 2005). Along with the socioeconomic development and emerging ecological environmental problems, global changes and ecosystem services are becoming the research hot topics. The relationships among LUC, climate change, and ecosystem services are interlaced and complex, in which temporal and spatial variations in human-induced LUC and climate variability can result in the difference of ecosystem services (Chen et al. 2013).

Natural ecosystem delivers a lot of benefits to human beings, and these benefits are known as ecosystem services. According to the Millennium Ecosystem Assessment (2005), these ecosystem services include provisioning services such as provision of food, water, timber, fiber, and genetic resources; regulating services such as the regulation of climate, floods, disease, and water quality as well as waste treatment; cultural services such as recreation, aesthetic enjoyment, and spiritual fulfillment; and supporting services such as soil formation, pollination, and nutrient cycling (Reid et al. 2005). Among all these services, supporting services and regulating services underpin the delivery of other service categories (Kumar 2010). What is more, there often exist trade-offs between different services when humans make management choices, which can change the type, magnitude, and relative mix of services provided by ecosystems (Rodriguez et al. 2006). However, people generally prefer provisioning and cultural services over regulating services (Carpenter et al. 2006) and thus tend to undervalue regulating services. Consequently, decision-makers often ignore these regulating services in ways that will seriously undermine the long-term existence of provisioning services (Kumar 2010). The regulating services provided by ecosystems are diverse, among which climate regulation is a final ecosystem service. The ecosystems regulate climate through biogeochemical and biogeophysical processes, as sources or sinks of greenhouse gases (GHGs) and as sources of aerosols all of which affect temperature and cloud formation (Bonan 2008; Fowler et al. 2009).

The processes involved in climate regulation include the following: (1) CO2 in the atmosphere was absorbed through photosynthesis; (2) evapotranspiration from soils and plants controls the amount of water vapor entering the atmosphere, thus regulating cloud formation and the radiative properties of the atmosphere; (3) the change of the albedo of different land surfaces can affect the climate; for example, the change in vegetation can have a cooling or heating effect on the surface climate and may affect precipitation; (4) the regulation of aerosols comes from soil erosion or vegetation through vegetation scavenging, which affects radiative heating of the atmosphere, surface albedo, and so forth (Bonan 2008).

There are many direct and indirect drivers that can affect the process of climate regulation services. For the systematic understanding of the effect mechanism and quantitative evaluation of the effects on climate regulation, it is important to clarify the major drivers and quantitatively analyze the effects induced by LUC and climate change on those drivers through both biogeochemical and biophysical processes. Comparatively, a direct driver more unequivocally influences ecosystem processes, while an indirect driver operates more diffusely by altering one or more direct drivers; that is, direct drivers have much more explicit effects on ecosystem processes (Nelson et al. 2005) and usually cause physical change that can be identified and monitored (Ash et al. 2008). As to climate regulation services, the indirect drivers mainly include demographic drivers (population growth and distribution, migration, ethnicity, etc.), economic drivers (economic growth and consumer choice, market force, industry size, globalization, etc.), and sociopolitical drivers (legislation, regulation, etc.). The direct drivers are listed in Table 1.1, among which land-use drivers are the most important in the ecosystem context and can be identified as the main drivers of climate regulation (Anderson-Teixeira et al. 2012), while over longer term, climate change will also have feedback to climate regulation services (Pete et al. 2011).

Table 1.1
Direct drivers of changes in climate regulation and their corresponding effects, adapted from (Pete et al. 2011)

Category of drivers

Change of drivers

Effect on climate regulation services

Habitat change: land and sea use

Productive area: expansion, conversion, abandonment (agriculture, forestry)

Affects carbon sinks and existing stores, greenhouse gas (GHG) emissions, albedo and evapotranspiration, shade and shelter

Mineral and aggregate extraction (peat)

Affects soil carbon stores, GHG emissions

Urbanization and artificial sealed surfaces

Affects soil carbon stores, albedo, shade, shelter, local temperatures, and humidity

Pollution and nutrient enrichment

Pollution emissions and deposition

Affects aerosol sources (soot)

Nutrient and chemical inputs

Affects GHG emissions

Harvest levels/resource consumption

Livestock stocking rates

Affects GHG emissions

Climate variability and change

Temperature and precipitation

Affects existing carbon stores, evapotranspiration

CO2 and ocean acidification

Affects existing carbon stores, GHG emissions, aerosol sources

Sea-level change

Affects existing carbon stores

In this study, we focus on choosing close researches to explore how LUC and climate change affect the climate regulation through biogeochemical and biogeophysical processes, respectively. Then, incorporating with the effects on climate regulation services, researches on impact assessment for human well-being were further revisited. This study firstly describes the climate regulation services and the needs and significance to study climate regulation services, then examines how LUC and climate change affect climate regulation services through review of the researches on major drivers, and after that outlines how changes of the role of ecosystem services in regulating climate affect human well-being through investigation of four major aspects (economic value, extreme weather, food security, and human health) that are closely related to human well-being. This review study intends to provide a reference for the future research on LUC, climate change, climate regulation services, and human well-being. The framework for the review about the effects induced by LUC and climate change on climate regulation and further impact assessment for human well-being is shown in Fig. 1.2.


Fig. 1.2
Framework for integrating land-use-induced effects on climate regulation services into impact assessment for human well-being

LUC-Induced Effects on Climate Regulation Services

As mentioned above, LUC plays an important role in climate regulation. Anthropogenic land use has been and will continue to be a major driver of the changes in climate system (Anderson-Teixeira et al. 2012). And there have been many observations and simulations revealing that LUC exerts effects on climate regulation services through biogeochemical and biogeophysical processes.

Cumulative Effects Through Biogeochemical Processes

In terms of biogeochemical processes, LUC mainly affects the climate regulation services through the emission and sequestration of GHGs, especially through altering CO2 flux. The total amount of carbon stored in terrestrial biosphere is an important factor in climate regulation (McGuire et al. 2001). Terrestrial ecosystems contribute to climate regulation primarily through carbon dynamics. Plants absorb CO2 through photosynthesis, storing carbon in vegetation and soils, and the carbon accumulated in soil and biomass represents a pool of carbon which is greater than the atmospheric carbon pool (Lal 2004a). Deforestation, forest degradation, and other land-use practices accounted for approximately 20 % of global anthropogenic CO2 emissions during the 1990s (IPCC 2007). When carbon is released from the earth during cultivation, deforestation, fire, and other land-use practices, it binds with other chemicals to form GHGs in the atmosphere and accelerates global climate change (Lal 2004b). The conservation of carbon sinks or pools is therefore important to mitigate GHGs levels. Thus, it is of great significance to investigate the effect of LUC on climate regulation through its biogeochemical process, that is, through the effects on the cycle of carbon. And LUC can change the release of carbon to the atmosphere mainly through the disturbance on terrestrial vegetation and soils.

Biogeochemical Process Related to Terrestrial Vegetation

Terrestrial vegetation is a large carbon sink which plays an important role in the global carbon cycle and is valued globally for the services it provides to society. Vegetation classifications have been related to climate variables and used in the assessment of possible global response to climate change (Smith et al. 1992). Olson et al. (1983) built up a computerized database to document the map of vegetation and corresponding carbon density for natural and modified complexes of ecosystems. The map provides a basis for making improved estimates of vegetation areas and carbon quantities (Olson et al. 1983), illustrating that different types of vegetation have various ability of carbon sequestration. LUC can have great effects on the structure of terrestrial vegetation (Reidsma et al. 2006); for example, once the forest is being converted to cultivated land, the biodiversity will decline. The distribution of sources and sinks of carbon among the world’s terrestrial ecosystem is uncertain, through deforestation, urbanization, expansion of cultivated land, and other land-use practices; LUC variously alter the land surface and species compositions and exert various effects on the carbon cycle.

It is relatively difficult to quantify the process that LUC affects the carbon sink or net source of terrestrial vegetation. Most researches developed and applied different models to record the carbon emission or sequestration resulted from LUC. Firstly, many researchers adopted empirical data to simulate carbon emission and sequestration. Houghton et al. (1983, 2000, 2003) calculated the carbon emission resulting from LUC and the potential for sequestering carbon of different land covers mainly based on the parameters set for each vegetation type in each regime (Houghton et al. 1983, 2000, 2003), and the model was called the “bookkeeping” terrestrial carbon model. Further, Houghton et al. (2001) documented a numeric data package that consists of annual estimates of the net flux of carbon between terrestrial ecosystems and the atmosphere resulting from deliberate LUC, especially forest clearing for agriculture and the harvest of wood for wood products or energy from 1850 through 1990 (Houghton et al. 2001

Only gold members can continue reading. Log In or Register to continue