The global urban population has risen from about 3% of the world’s total population in the year 1800 (Brunn and Williams, 1983) to about 47% by 2000 (UN, 2004). In 2007, for the first time in history, more than half of the world’s population was living in cities, and the trend is expected to continue.
Megacities, often defined as cities with over 10 million people, have grown from only two in the 1950s to over 20 today. With the development of megacities there has been a concentration of anthropogenic pollutant emissions and land-use change that has large environmental implications from local to regional scales. Rapid economic growth in megacities is often not followed by an equally rapid growth in infrastructure such as road construction or public transportation systems, leading to inefficient traffic flow and enhanced air pollution. For instance, Gurjar and Lelieveld (2005) point out that the per capita CO emissions in megacities tend to be higher than in densely populated countries.
Furthermore, a recent study of Asian megacities found that they cover less than 2% of the land area, hold more than 30% of the population and produce ~10% of the anthropogenic gas and aerosol emissions. These numbers mean that large fractions of the population are exposed to disproportionately adverse health impacts in megacities (Guttikunda et al., 2005). The increasing emissions from the emerging and evolving megacities as well as changes in the emission patterns increase the severity of several environmental problems, in particular in relation to air pollution, climate change, water and soil resources. It has become urgent to assess and address this broad problem in the science and policy context.
CityZen (megaCITY Zoom for the ENvironment) is a three year European Commission funded project finishing in 2011. The focus of the project has been to address the complex interactions of megacities within the air quality and climate change context using state-of-the-art models, in-situ measurements, and satellite observations. The main objectives of CityZen, of which selected preliminary results will be presented, are:
• Quantify and understand current air pollution distribution and development in and around selected megacities/hot spot regions, including the interaction across the different spatial scales.
• Estimate the future impact of emission changes with a focus on the effect of rapid growth in the population of megacities/hot spots and the increasing background of pollutants, with a concentration on ozone, particulate matter, and their precursors.
• Estimate how megacities/hot spots influence climate change.
• Estimate how megacities are responding to climate forcing which can influence transport patterns, chemical oxidation and biogenic emissions, with a concentration on biogenic volatile organic compounds.
• Study mitigation options to limit adverse human health effects and climate impact.
• Develop tools to estimate interactions between different spatial scales (megacities to global).
• Communicate the project results to policy makers providing a scientific underpinning of policy work.
Motivation for Study Regions From a human health perspective in the European Union (EU), the pollutants of greatest concern are ground level ozone and particulate matter (PM), specifically fine PM (PM2.5) (EU, 2005). Ozone is a secondary air pollutant, which means that it is formed in the atmosphere and has no significant emission sources. Ground level ozone is formed from photochemically driven reactions of primarily nitrogen oxides (NOx = NO2 + NO), volatile organic compounds (VOCs), and carbon monoxide (CO). These ozone precursors are emitted in significant amounts from the transportation sector, as well as power generation, and industrial sources which are significant in urban areas. Ozone is of particular interest on varying spatial scales because of the time it takes to form in the atmosphere, which can thereby have a significant impact not only on the urban area, but on the surrounding region as well. In addition to the adverse impacts on human health, ozone can negatively impact vegetation, resulting in adverse consequences for crop yields.
Anthropogenic sources of fine particulate matter are emitted primarily from combustion sources, including the transportation sector, power generation, and industry. The smaller size fraction of PM is of greater concern for human health than the larger size fraction (PM10) because of its ability to penetrate deeper into the lungs. The health effects of air pollution in the EU were estimated as a loss in statistical life expectancy of more than 8 months due to PM2.5 in air, which corresponds to 3.6 million life years lost annually (EU, 2005). For ozone, roughly 21,000 cases of hastened mortality in 2020 were estimated for the EU. These adverse health effects are of particular consequence to more vulnerable parts of the population, such as children, the elderly, and those suffering from asthma and cardiovascular diseases (EU, 2005).
In CityZen an emissions hot spot approach was used with the main megacity/hot spot areas focused on the Po Valley in Italy, the BeNeLux region, London, Paris, and the Eastern Mediterranean including Athens, Istanbul, and Cairo (Figure 1). These European megacities are contrasted with the Pearl River Delta in China, including Guangzhou and Hong Kong. The Focus Regions The Eastern Mediterranean hotspot region includes Istanbul as a megacity and Athens -extended area, about 5.5 million people, and Cairo metropolitan area which is a megacity with more than 16 million inhabitants. The Eastern Mediterranean is a cross road of air masses where man-made pollution meets with natural emissions. These emissions include nitrogen oxides, carbon monoxide, volatile organic compounds, as well as particulate matter. The main pollution sources include the industrial sector, transportation, and domestic heating, as well as transport from the continental part of Europe, Balkans and the Black Sea. Local and regional forest fires and agricultural/biomass burning emissions are also affecting the area during the dry season. Spots of intense burning activity are found north of the Black Sea and combined with the dominance of the northerly winds over the Aegean Sea, contribute significantly to pollution build up during summer. Natural emissions from semi-arid North African regions (e.g. Sahara desert), from the vegetation that surrounds the Mediterranean Sea and from the Mediterranean Sea itself also affect the area. Southerly winds from Africa mobilises the transport of air masses with high loads of dust and low levels of NOx and O3 over the Eastern Mediterranean. This air flow pattern is most frequent during the transition periods (spring and autumn). Air masses transported from the Atlantic Ocean atmosphere and within the upper troposphere from Asia can also reach the area under certain air flow conditions. The Po Valley hotspot region also affects the East Mediterranean basin at times. Several large, continuously growing, agglomerations are located inside the Eastern Mediterranean hotspot region or close to it; Athens, Istanbul and Cairo have grown dramatically over the past 10 years. Pearl River Delta (PRD) is an area with one of the fastest economic growth rates in China. Urbanization in the PRD is characterized by city clusters with two megacities (Guangzhou and Hong Kong) and many medium-sized to small cities linked by dense highways. Until around 1985, the PRD had been mainly dominated by farms and small rural villages, but after the economy was reformed and opened, the Pearl River Delta has become the most economically dynamic region of China with an average rate of GDP growth exceeding 16% annually during the period 1980-2000. The expansion of the economy in this region causes ever higher demands for energy, mobility and communications.
As a consequence, smog from coal burning and traffic exhaust together cause serious photochemical smog and particulate pollution problems from urban to regional scale. Atmospheric visibility has deteriorated with less blue sky each year owing to high concentrations of O3 and fine particles. Especially under stagnant conditions the levels of PM2.5 and ozone represent a serious impact on human health. The radiative forcing by aerosols and black carbon attain high levels and serious impact on climate is to be expected. Atmospheric chemistry is complicated owing to interactions between primary emissions and photochemical processes, between gaseous and aerosol phase, and between local and regional air pollution. Transformation and transport of air pollutants show rather unique characteristics under such conditions of high concentrations of primary and secondary pollutants. The possible impact on regional air quality and climate change is a major concern of national government as well as the global community. Much of the area is frequently covered with brown smog. This has a strong effect on the pollution levels in Hong Kong. Long-term records of space observations have shown that the increases in pollutant amounts over China have been much larger than what was originally expected from emission estimates (Richter et al., 2005).
Technical Underpinning of Policy – Preliminary Scientific Findings One component of the CityZen project is to work with national authorities, intergovernmental organisations, and EU institutions to prepare for and whenever possible provide technical underpinning for policy. While individual project partners work within and communicate with the policy-forming side of their organizations, CityZen has and will continue to produce documents containing concise policy-relevant summaries of the scientific conclusions. Preliminary messages with relevance for air quality or climate policy formation have already been issued.
A selection of the preliminary scientific findings is presented here:
• Analysis of the trends in air pollutants over the past decade, mainly ozone precursor species and aerosols, from satellite and ground-based observations in the CityZen regions have generally found decreasing trends in Western Europe and increasing trends elsewhere. A study of NOx trends, as observed from satellites from 1996 to 2008 in twelve megacity areas in Europe and the Middle East (London, Paris, Essen, Berlin, Milan, Madrid, Barcelona, St. Petersburg, Moscow, Istanbul, Bagdad, and Tehran) found that strong negative trends tend to dominate in Western Europe, while positive trends dominate outside of Western Europe (Konovalov et al., 2010). The modelled linear trends in NOx emissions ranged from -5.2 % per year (Berlin) to +4.5 % per year (Tehran). The exceptions to this were Barcelona and Madrid which showed increasing NOx trends (Konovalov et al., 2010). A subset of European air quality monitoring stations was derived from the comprehensive Airbase repository in order to extend existing trend assessment work (usually based on remote monitoring sites) to urban areas. This subset confirmed the findings above regarding downward trends of NO2 concentrations in Western Europe, also showing that it was accompanied by mild increases of O3 concentrations in NOx-saturated areas – raising an important point for consideration when shaping future air quality management strategies. The same data was mined to assess PM10 trends, finding significant downward trends at 37% and significant upward trends at 16% of the European monitoring sites investigated. (Summarized from (Colette et al., 2011)). Ground-based measurements from a monitoring site within London showed similarly decreasing trends from 1998 to 2009 for non-methane hydrocarbons (NMHCs) and CO, but at a much faster pace in some cases. Ethane and propane showed the slowest rate of decrease, -4% per year and -3% per year, respectively, while the remaining NMHCs (C2-C8) decreased between -12% per year and -26% per year. Carbon monoxide showed a trend of -12 % per year (von Schneidemesser et al., 2010). Satellite observations of NOx and aerosols have shown increasing levels of pollution in the Pearl River Delta region of China from 1996 to 2010. Similar observations of sulfur dioxide levels in east central China were increasing up to 2007, but have been decreasing since then. This indicates that measures taken to remove sulfur from the emissions of coal fired power plants have been effective (personal communication, A. Richter).
• Trends in PM10 and NO2 over Western Europe for the past decade showed downward trends, which were reproduced well by a number of models. A coordinated modelling experiment was designed to assess the capacity of state-of-the-art chemistry-transport tools to reproduce air quality trends and inter-annual variability. Four regional and two global models simulated the past 10 years over Europe and beyond, some of them duplicating the simulation with constant emissions of pollutants to assess the respective role of meteorological and anthropogenic variability. The models consistently found downward trends of NO2 and PM10 over most of Western Europe, in agreement surface monitoring stations giving confidence in their ability to reproduce air quality trends. The much milder trend of O3 were more challenging to capture, as well as its year-to-year variation that was shown to be dominated by the meteorological variability. (Summarized from (Colette et al., 2011)).
• The East Mediterranean region has significant air pollution, often exceeding air quality targets, owing to a number of factors, including high background levels, significant natural emissions, climate, and location of the Mediterranean with respect to local and regional emission sources. Various studies conducted in the Eastern Mediterranean, including Athens and Istanbul have observed significant contributions of natural sources, specifically dust from the Sahara and other arid regions in the area as well as maritime emissions such as sea-salt. These natural contributions can account for one-quarter of PM mass to over half of PM mass during dust-transport events, even in urban areas (Hatzianastassiou et al., 2009; Kocak et al., 2010; Theodosi et al., 2011; Theodosi et al., 2010). Certain meteorological conditions can contribute to elevated background levels of pollution to the region, as recent measurements (2005-2006) of PM at background stations in the Eastern Mediterranean have shown (Theodosi et al., 2011). In addition to the significant contribution from regional background levels and natural sources, local sources of pollution, such as traffic, domestic heating, and industry are particularly important for determining aerosol levels in the urban areas and megacities (Gerasopoulos et al., 2011; Hatzianastassiou et al., 2009; Kanakidou et al., 2011; Kocak et al., 2010; Theodosi et al., 2011; Theodosi et al., 2010).The Mediterranean is located at a crossroad of air masses, being influenced by pollution transported from Europe, Asia, and Africa, as well as more local sources (Gerasopoulos et al., 2011; Kanakidou et al., 2011). Figure 2 shows enhanced levels of pollution over the greater East Mediterranean region as observed by satellites. Finally, the climate of the Mediterranean is such that solar radiation is intense year-round, driving photochemical reactions and negatively affecting air pollution, increasing the likelihood of photochemical pollution episodes, especially, but not only, in summer (Kanakidou et al., 2011). In addition, temperature changes due to meteorological variability and climate change are expected to significantly impact atmospheric composition (Im et al., 2011). These characteristics contributing to air quality in the Eastern Mediterranean mean that aerosol and ozone air quality limits are frequently exceeded, especially in the regions’ megacities, and a variety of influences need to be considered when forming air quality policy to try and address these issues.
• A London case study found that isoprene (a proxy for biogenic VOCs) still plays a relatively small role in overall VOC reactivity, and thereby ozone formation. The exception to this is periods of high temperature and sunlight, which increases biogenic emissions significantly (e.g., summer 2003 heat wave). While significant decreases in anthropogenic VOCs were observed over the past decade, likely owing to effective regulation of emissions, isoprene has not seen a rise in importance relative to the anthropogenic VOC fraction (von Schneidemesser et al., 2010). Furthermore, a significant fraction of the isoprene in urban areas likely originates from anthropogenic sources. To reliably assess the influence of biogenic emissions in a region it may be necessary to include the measurement of monoterpene species in routine monitoring, as isoprene, which is frequently used as a proxy for biogenic emissions, may not be sufficient. These results may have implications for urban air quality policies and monitoring practices, as well as emission inventories used in modelling. (Summarized from (von Schneidemesser et al., under review)).
• A modelling study of ozone formation in the Pearl River Delta region of China found that ozone formation is likely VOC-limited in the urban, inland areas and NOx-limited in the more rural areas with a predominance of aged emissions (Wang et al., 2010). The Pearl River Delta region of China has significant photochemical pollution episodes. Emissions of nitrogen oxides in the PRD area are dominated by mobile sources (47%) and power generation (39%); the three largest sources of VOCs are mobile sources (38%), evaporation losses of solvents and petroleum (24%) and biogenic sources (23%), with the largest concentration of emissions over the urban areas in the region. The ozone response to reductions of anthropogenic VOCs and NOx, separately and simultaneously revealed significant spatial differences in VOC- and NOx-limited conditions. In the urban areas, less efficient ozone production resulted from intense NOx emissions which reacted with and thereby suppressed ozone concentrations. As emissions were transported out of the urban areas and concentrations of NOx were lower, ozone production efficiency increased. These varying regimes of ozone production efficiency over the PRD region are shown in Figure 3. Additionally, synoptic weather conditions in the PRD significantly influence the formation and distribution of ozone. (Summarized from (Wang et al., 2010)). The different regimes and their effect on ozone production efficiency should be considered when forming and implementing policy for emission reductions to achieve the largest benefits.
• High resolution of emissions is important on a local/megacity scale for understanding pollutant production and exposure, but less important for climate studies that aim at understanding changes in pollutant production over a regional to global scale. Urban- to global-scale models are used to simulate and improve understanding of ozone production and distribution. However, urban scale models are typically used to assess ozone in terms of air quality, while global scale models focus on ozone in terms of climate change and its role as a greenhouse gas. As the emissions from megacities leading to ozone production are substantial, and spatial distribution of emissions are not even, this study aimed at understanding and quantifying the inaccuracies that arise when ozone production is modelled on a coarse scale (e.g. in a global chemistry-transport model). This was accomplished by applying a regional model using finer and coarser resolution of emissions over large urban areas. The main difference resulting from the varying resolution of the emissions was that finer resolution was able to capture areas of high NOx in pollution plumes which caused suppressed ozone levels, whereas in the coarse resolution few grid boxes reached such high concentrations of NOx. This resulted in a small enhancement of ozone production in the coarser grid models in most of the cases that were studies. Overall these differences were small and the coarser resolution was deemed sufficient for climate studies, whereas the high resolution would be more important for local air pollution studies. (Summarized from (Hodnebrog et al., 2011)). An established approach to focus on urban scale processes in regional models is two-way nesting. Two-way nesting involves simulating several sub-domains and their interactions, simultaneously, however this technique is computationally expensive. In CityZen, the impact of megacities on the regional scale was addressed by innovative chemistry-transport modelling using scale bridging approaches: grid stretching (Siour et al., under review) and upwards nudging (Maurizi et al., 2011). Grid stretching involves a single grid with variable horizontal resolution that permits focusing on selected areas, while upwards nudging uses data from high resolution model runs and forces this data into low resolution areas to improve concentration fields. By allowing a feedback of the inner (high resolution) grid area to the larger, coarser grid scale, these strategies offer an alternative to two-way nesting. Both approaches confirmed that an improved representation of megacities was crucial to capture air quality in their vicinity although the added value was limited compared to the classical meso-scale single-nested approach regarding air pollution export at the larger (regional) scale.
• Under climate conditions expected for 2030, model simulations predict increased ozone concentrations, however, projected emission reductions will more than compensate for this and therefore most areas have predicted ozone reductions (Figure 4). Until 2030, according to model results, the predicted effect of climate change on surface ozone is an increase in almost all of Europe (not shown). The only exceptions are the Northwest of the British Isles and the Northwest of Russia. However, when projected reductions in ozone precursor emissions are taken into account (Figure 4), surface ozone is assumed to decrease, i.e. the decrease due to emission reductions overwhelms the increase due to climate change. (For the projected reductions see (Benedictow et al., 2010)). The only exceptions are the highly polluted areas such as London, BeNeLux, parts of the Po Valley and Moscow, where the reduction in titration following the NOx emission reductions will lead to further increases in ozone. This finding is another illustration that air quality improvements in ozone will remain limited as long as this area remains saturated in NOx emissions.
• Under climate conditions expected for 2030, model simulations predict an overall reduction in PM2.5 concentrations, as projected reductions in anthropogenic emissions of aerosols and aerosol precursors outweigh indirect increases in concentrations due to climate change (Figure 4). In the case of particulate matter (PM2.5), climate change will lead to increases over large parts of Western and Southern Europe as well as the Mediterranean, mainly because of changes in precipitation and wash-out. Reductions are modelled in Eastern Europe. Similarly to the ozone case, the projected reductions in anthropogenic emissions (aerosols and aerosol precursors) overcompensate for the effect from climate change (Benedictow et al., 2010). The model predicts an overall reduction in PM2.5 concentrations when both climate change and emission reductions are taken into account. In summary, CityZen has made significant progress into understanding and quantifying the impacts of megacities in Europe, the near East, and China on air quality, and the impacts of climate change on air quality in densely populated areas. The measurements and models in combination have shown that there are strong regional characteristics affecting the megacity regions focused on in CityZen. The East Mediterranean, Europe, and China hot spot regions all have vastly different meteorology and dominating influences, and yet face similar air quality and climate change challenges in the future. Understanding these differences and what tools are necessary to do so will provide important input for future planning and decision-making for air quality and climate policy. As the project finishes, targeted special topics summaries will be issued that provide focused messages summarizing the findings from CityZen that are most relevant for scientific policy underpinning.
References
Benedictow, A., Berge, H., Fagerli, H., Gauss, M., Jonson, J.E., Nyiri, A., Simpson, D., Tsyro, S., Valdebenito, A., Shamsudheen, S.V., Wind, P., Aas, W., Hjellbrekke, A.-G., Mareckova, K., Wankmueller, R., Iversen, T., Kirkevag, A., Seland, Ø., Haugen, J.E., Mills, G., 2010.
Transboundary Acidification, Eutrophication and Ground Level Ozone in Europe in 2008. Meteorologisk Intitutt. Colette, A., Granier, C., Hodnebrog, Ø., Jakobs, H., Maurizi, A., Nyiri, A., Bessagnet, B., D'Angiola, A., D'Isidoro, M., Gauss, M., Meleux, F., Memmesheimer, M., Mieville, A., Rouïl, L., Russo, F., Solberg, S., Stordal, F., Tampieri, F., 2011.
Air quality trends in Europe over the past decade: a first multi-model assessment. Atmospheric Chemistry and Physics Discussions 11, 19029-19087. EU, 2005.
Communication from the Commission to the Council and the European Parliment: Thematic Strategy on air pollution, in: Communities, C.o.t.E. (Ed.), Brussels. Gerasopoulos, E., Amiridis, V., Kazadzis, S., Kokkalis, P., Eleftheratos, K., Andreae, M.O., Andreae, T.W., El-Askary, H., Zerefos, C.S., 2011.
Three-year ground based measurements of aerosol optical depth over the Eastern Mediterranean: the urban environment of Athens. Atmospheric Chemistry and Physics 11, 2145-2159. Gurjar, B.R., Lelieveld, J., 2005.
New Directions: Megacities and global change. Atmospheric Environment 39, 391-393. Guttikunda, S.K., Tang, Y., Charmichael, G.R., Kurata, G., Pan, L., Streets, D.G., Woo, J.-H., Thongboonchoo, N., Fried, A., 2005.
Impacts of Asian megacity emissions on regional air quality during spring 2001. Journal of Geophysical Research 110, D20301. Hatzianastassiou, N., Gkikas, A., Mihalopoulos, N., Torres, O., Katsoulis, B.D., 2009.
Natural versus anthropogenic aerosols in the eastern Mediterranean basin derived from multiyear TOMS and MODIS satellite data. Journal of Geophysical Research 114, D24202. Hodnebrog, O., Stordal, F., Berntsen, T.K., 2011.
Does the resolution of megacity emissions impact large scale ozone? Atmospheric Environment in press. Im, U., Markakis, K., Poupkou, A., Melas, D., Unal, A., Gerasopoulos, E., Daskalakis, N., Kindap, T., Kanakidou, M., 2011.
The impact of temperature changes on summer time ozone and its precursors in the Eastern Mediterranean. Atmospheric Chemistry and Physics 11, 3847-3864. Kanakidou, M., Mihalopoulos, N., Kindap, T., Im, U., Vrekoussis, M., Gerasopoulos, E., Dermitzaki, E., Unal, A., Koçak, M., Markakis, K., Melas, D., Kouvarakis, G., Youssef, A.F., Richter, A., Hatzianastassiou, N., Hilboll, A., Ebojie, F., Wittrock, F., Savigny, C.v., Burrows, J.P., Ladstaetter-Weissenmayer, A., Moubasher, H., 2011.
Megacities as hot spots of air pollution in the East Mediterranean. Atmospheric Environment 45, 1223-1235. Kocak, M., Theodosi, C., Zarmpas, P., Im, U., Bougiatioti, A., Yenigun, O., Mihalopoulos, N., 2010.
Particulate matter (PM10) in Istanbul: Origin, source areas and potential impact on surrounding regions. Atmospheric Environment. Konovalov, I.B., Beekmann, M., Richter, A., Burrows, J.P., Hilboll, A., 2010.
Multi-annual changes of NOx emissions in megacity regions: nonlinear trend analysis of satellite measurement based estimates. Atmospheric Chemistry and Physics 10, 8481-8498. Maurizi, A., Russo, F., D’Isidoro, M., Tampieri, F., 2011.
Nudging technique for scale bridging in air quality/climate atmospheric composition modelling. Atmospheric Chemistry and Physics Discussions 11, 17177-17199. Richter, A., Burrows, J.P., Nuess, H., Granier, C., Niemeier, U., 2005.
Increase in tropospheric nitrogen dioxide over China observed from space. Nature 437, 129-132. Siour, G., Colette, A., Menut, L., Bessagnet, B., Coll, I., Meleux, F., under review. Bridging the scales in an Eulerian air quality model to assess megacity export of pollution. Environmental Modelling and Software. Theodosi, C., Grivas, G., Zarmpas, P., Chaloulakou, A., Mihalopoulos, N., 2011.
Mass and chemical composition of size-segregated aerosols (PM1, PM2.5, PM10) over Athens, Greece: local versus regional sources. Atmospheric Chemistry and Physics Discussions 11, 7659-7691. Theodosi, C., Im, U., Bougiatioti, A., Zarmpas, P., Yenigun, O., Mihalopoulos, N., 2010.
Aerosol chemical composition over Istanbul. Science of the Total Environment 408, 2482-2491. von Schneidemesser, E., Monks, P.S., Gros, V., Gauduin, J., Sanchez, O., under review.
How Important is (Biogenic) Isoprene in an Urban Environment? - A Study in London and Paris. Geophysical Research Letters. von Schneidemesser, E., Monks, P.S., Plass-Duelmer, C., 2010.
Global Comparison of VOC and CO Observations in Urban Areas. Atmospheric Environment 44, 5053-5064. Wang, X., Zhang, Y., Hu, Y., Zhou, W., Lu, K., Zhong, L., Zeng, L., Shao, M., Hu, M., Russell, A.G., 2010.
Process analysis and sensitivity study of regional ozone formaiont over the Pearl River Delta, China, during the PRIDE-PRD2004 campaign using the Community Multiscale Air Quality modeling system. Atmospheric Chemistry and Physics 10, 4423-4437.
Acker, J.G., Leptoukh, G., 2007. Online analysis enhances use of NASA earth science data. Eos, Transactions, American Geophysical Union 88 (2), 14e17.
|
is Published by 
Figure 1. Population density map with the CityZen hotspot areas circled. Credit: NASA, Visible Earth (http://visibleearth.nasa.gov/), National Center for Geographic Information and Analysis (NCGIA).