Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 29
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report 3 Challenges and Potential for Applying Land Remote Sensing to Human Welfare Though there have been great advances in the use of remotely sensed data in the Earth sciences and global climate research in the past several decades, the potential for applying remotely sensed data to a range of human welfare issues is largely undefined and unrealized. What does remotely sensed data offer that other capabilities do not? How can realizing the potential enable decisions that benefit human health and welfare? What can be done to overcome the barriers to the use of remotely sensed data for human welfare purposes? Workshop participants again assembled into two groups to discuss these questions as they relate to food security and human health issues, respectively. Workshop participants noted that as in other areas of science, the integration of knowledge gained from remotely sensed data into decisions on human welfare can be categorized into four separate processes: observing, explaining, projecting (forecasting), and applying in practice. Remote sensing applications are usually associated with observing and then, to a lesser extent, with explaining and projecting. Remote sensing provides a means to indirectly observe patterns or changes, which then informs explanations of causal relationships and processes. Projections are based on observations, understanding of processes, and assumptions about the future. Modeling tools developed for making projections are tested and refined by further observations. The modeling tools can then be used to aid decision makers in responding to crises or implementing longer-term development strategies. Creating a thread linking observation, explanation, and projection to application is vital to providing appropriate information to the domain of decision makers.
OCR for page 30
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Chapter 2 of this report offers examples of how the use of remotely sensed data can contribute to human health and welfare. This chapter summarizes workshop discussions regarding opportunities and challenges in the application of remotely sensed data of attributes of the land surface such as land cover, human infrastructure, productivity of vegetation, and seasonality for improving food security and human health. Workshop participants recognized that environmental and socioeconomic information provided by remote sensing and other sources constitutes only one aspect of effective decision making. The political, economic, and institutional setting defines the framework in which such information can be used. Decision makers need relevant information to make informed decisions that will have significance and impact and also need robust institutional capabilities to be able to support and carry out their decisions. While good information and institutional capabilities are both critical for effective decision making, it is often the lack of the latter that inhibits good decision making from being actualized. THE POTENTIAL OF REMOTE SENSING APPLICATIONS FOR FOOD SECURITY Although substantial donor resources are dedicated to providing food aid, the ability to monitor food availability and predict food shortages is of equal—or arguably greater—importance in promoting food security. Food security issues can be divided into three general elements: food availability, accessibility, and utilization. Integrating remote sensing information into decision-making networks can improve the capacity of decision makers to make effective choices in each of these domains. Remote sensing technologies provide seasonal information that is cheaper, more timely, and available over larger areas than traditional sources of information on food accessibility. Greater investments in remote sensing infrastructures could aid policy makers in forecasting future food availability and accessibility and to plan appropriately, perhaps by instituting multigenerational development efforts. The production and delivery of food resources to populations in need involves interactions between a complex array of social and environmental factors. Decision makers must understand where food shortages are likely to occur and be aware of the location and accessibility of surplus food resources. Remote sensing identifies vulnerabilities based on environmental variability and can enhance land-based environmental data. Incorporating information from remotely sensed data, in combination with administrative and agricultural records over long periods of time, can allow decision makers to identify areas more statistically prone to food crises and to identify the physical accessibility and quality of needed resources. In order to lessen the severity of food crises, decision makers
OCR for page 31
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report must be aware of factors that contribute to them, including general human and livestock health and the quality and availability of clean water supplies. Strategically deciding to fix wells and improve human and livestock health prior to a crisis may ultimately be of greater long-term benefit than supplying food, because health and income sources will then be more resistant to crises and require less food aid. A key challenge in the effective use of remotely sensed data is getting the appropriate data products into the hands of the appropriate decision makers, in time frames reasonable for action. Although there are examples of government agencies (such as the U.S. Geological Survey, Environmental Protection Agency, and regional multi-government agency groups such as the South African Development Community), international agencies (such as the United Nations Food and Agriculture Organization), and nongovernmental organizations (NGOs) (such as conservation NGOs including the World Wildlife Fund and Conservation International) developing the required technical knowledge to take advantage of remote sensing technologies, there is generally insufficient communication between the community interpreting remote sensing data and the policy-making community. The challenge is to develop the capacity for these groups to communicate so that the remote sensing community understands the needs of policy makers and policy makers understand the capabilities of remote sensing. To help decision makers comprehend the potential of remotely sensed data applications, data must be provided in the form of knowledge delivery and as a tool for decision making, rather than as a list of observations or prescriptions. The integration of data is key: when combined with socioeconomic data, remote sensing data can be useful for direct applications to decisions about human welfare. The Famine Early Warning System Network (FEWS NET), for example, has monthly briefings with the U.S. Department of Agriculture and other organizations and distributes a two-page executive overview describing the status of food security in Africa several times a month (see Appendix C). Such communication distills data analysis down to concise points, identifying areas of risk on a country-level map for high-level decision makers. By learning what information decision makers need, FEWS NET can translate remotely sensed data into information that contributes directly to effective planning for potential food shortages. To date, there has been no indication of the long-term changes and practices due to FEWS NET activities. An evaluation of programs such as FEWS NET is needed, since the effectiveness of its methods has not been fully analyzed. To process and provide the appropriate remotely sensed data and information, infrastructures created by groups such as FEWS NET must consist of teams of remote sensing specialists, ground receiving stations,
OCR for page 32
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report and web-based data archives. FEWS NET’s infrastructure allows fairly effective communication among specific agencies, providing ample opportunity for policy makers and the remote sensing community to communicate the needs and abilities of each. In many cases, however, decision makers are unable to react to what has been learned from remotely sensed data. The crisis may already exist and procurement cycles may be out of phase with the timing of data used for policy decisions. Remote sensing infrastructures do not exist, and expertise is not available in many at-risk areas to take advantage of the current state of the technology of remote sensing applications. NGOs often fill this gap by providing critical expertise, particularly in developing countries. Many NGOs, intergovernmental organizations, and aid organizations are international in scope, and often have excellent in-house technical resources (such as the United Nations Food and Agriculture Organization). In developing countries, national universities can play a critical role in developing local remote sensing infrastructures by disseminating remote sensing resources and training future generations of remote sensing specialists. Many universities in less developed nations have connections to U.S. land-grant institutions that can provide them with means to acquire useful and necessary skills to tackle local problems. Despite an awareness of the value of remote sensing applications, however, there are often only limited resources dedicated within organizations focused on remote sensing applications. To realize the full potential of remote sensing for food security issues, integrated approaches are necessary that relate the real-time connections between food, water, and health. The importance of remote sensing applications to food security goes beyond disaster relief. The scientific community can be most useful to decision makers by integrating the biophysical domain with social data. Livelihood analysis, as done by FEWS NET, is an example of socially informed remote sensing analysis and the benefits of cross-disciplinary collaborations. Remote sensing, natural, and social scientists are beginning to develop approaches to integrate the natural and social sciences with remotely sensed data and can work together with decision makers to apply remote sensing information to decisions about human welfare. THE POTENTIAL OF REMOTE SENSING APPLICATIONS TO HUMAN HEALTH Remotely sensed data provide a spatial perspective on human health issues not typically incorporated into human health research and applications. Remotely sensed data, as applied to human health and welfare, can assist in taking into account multiple factors affecting health, such as food production,
OCR for page 33
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report availability, and distribution; environmental health hazards; contagious and infectious diseases; chronic health issues; and health delivery. All of these factors have traditionally been considered separately by a variety of organizations. Health professionals do not typically observe the human land uses and ecological conditions affecting human health from the viewpoint of the remote sensing satellite. For health professionals, who often view issues in terms of point estimates or community averages, the visual and spatial perspective from remote sensing fosters a more integrated approach. Remotely sensed data, in combination with other data, can provide spatial information on environmental conditions for understanding distributions of water-borne disease, air quality, soil, and vegetation as they influence community health and livestock. Remotely sensed data also provide spatial information on land use and infrastructure, which aids in determining where people live, where vulnerable populations live, the distribution of urban populations, and the quality of roads and other infrastructure for health care delivery. Interdisciplinary and international collaboration are needed between remote sensing scientists, ecologists, and human health scientists to realize the full potential of remote sensing applications. The successful application of remotely sensed data to public health issues depends on several factors, including the need for cooperation between the remote sensing and public health communities. From the remote sensing side, successful application depends on the following: Identifying what parameters on the Earth’s surface need to be measured remotely and for use in health applications; Determining the relevant variables that can be extracted from existing remote sensing satellites, such as human infrastructure and habitats for disease vectors; Developing multi-resolution sensors at multiple spatial and temporal resolutions to monitor a range of health phenomena, especially sensors that are able to detect conditions in urban areas with complex mixtures of vegetation, buildings, and roads; Educating local health officials on possible contributions of remotely sensed data; and most importantly, Maintaining the continuity of current satellite coverage so that change of environmental conditions and their effects on health can be monitored over extended time frames. It would be helpful if collected data, such as those listed in Table 3-1, were integrated with socioeconomic data on population distributions and livelihoods to realize the potential for human health applications. If data directly and indirectly measure disease, the data can be used
OCR for page 34
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report TABLE 3-1 Environmental Conditions and Change Requiring Monitoring by Multiple Types of Remote Sensing (e.g., optical, radar, microwave) Condition Observations Benefits Examples of Remote Sensing Technologies and Existing Sensors Water Water quality (e.g., temperature, oxygen content) Water availability Water locations and types (e.g., wetlands, lakes) Rainfall Monitor conditions conducive to waterborne disease growth or migration (worms, flu, meningitis, cholera, malaria, West Nile virus, AIDS); wetland mapping Radar, multi-spectral optical (Landsat) Air and atmosphere Ozone measurements Particulates Heat and temperature UV measurements Wind dynamics Dust movements Air quality, atmospheric chemistry, climate change—monitoring these conditions allows for indirect measurement of diseases such as asthma Thermal reflectances (MOPITT) Soil and vegetation Soil moisture Vegetation types Vegetation productivity Habitats for disease vectors Multi-temporal, multi-spectral optical (MODIS, Landsat) Land use and land cover Land cover Livestock NDVI Cropland extent Soil, water, and livestock interactions; land-sea interface; detection of floodplains, ice cover Multi-temporal, multi-spectral optical (MODIS, Landsat)
OCR for page 35
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Condition Observations Benefits Examples of Remote Sensing Technologies and Existing Sensors Infrastructure Roads and transportation Water access Sewers Communications Waste disposal Urban population distributions at high resolution Disease vector tracking; improving health service response in times of emergency; developing GIS data layers for modeling housing, land cover, etc.; high-resolution population distributions that can assist in health issues associated with infrastructure (i.e., obesity as related to infrastructure); understanding of teleconnections Very high resolution optical (IKONOS, QuickBird) NOTE: GIS = geographic information systems; NODIS = Moderate Resolution Imaging Spectroradiometer; MOPITT = Measurement of Pollution in the Troposphere; NDVI = Normalized Difference Vegetation Index. to map diseases, estimate the burden of disease (the suffering of the society beyond that directly caused by the disease), and target intervention strategies. From the health researcher’s point of view, there are several challenges that bar the regular application of remotely sensed data in the health arena: Remotely sensed data are expensive; affordable data access is key to bringing the technology to health applications. There are few in situ ground truth data accompanying remote sensing technology, which will inhibit the use of the technology. In situ data are needed to verify analyses of data derived from remote sensing and to collect other data that cannot be detected remotely. Health research professionals are not typically trained to use the tools required to analyze remotely sensed data. Privacy concerns inhibit the sharing of health data by the health care community. In particular, lack of locational information on disease cases inhibits spatial analyses using remote sensing.
OCR for page 36
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Overcoming these obstacles will greatly enhance the toolbox of health researchers by allowing them to more easily understand the relationships among infrastructure, the physical properties of land and water, health statistics, and access to health care delivery. To realize and maximize the potential of applying remotely sensed data to decisions about health and health care, the issues of the remote sensing and health research communities can be solved simultaneously, with cooperation between the two. Workshop participants noted that achievable, short-term (five-year) priorities include improving the understanding of ecological processes in health (e.g., from vector-borne diseases including hantavirus, cholera, and West Nile virus) and learning how to fuse remotely sensed data with traditional land-based data (e.g., settlement and population data). Most workshop participants observed that limitations in data access, data sharing, and the continuity of observations have to be addressed. Long-term priorities are to develop an end-to-end system to support health applications and multi-scale, multi-temporal data for surveillance, monitoring, and prediction. REALIZING THE POTENTIAL OF REMOTE SENSING FOR HUMAN WELFARE APPLICATIONS: COMMON THEMES FROM HUMAN HEALTH AND FOOD SECURITY Workshop discussions highlighted that the use of remote sensing data for human health applications is not as advanced as its application for food security issues. The interdisciplinary collaborations are not as firmly established as, for example, those between the U.S. Department of Agriculture and groups monitoring agricultural yields. Despite this difference, workshop participants identified common themes that are applicable to both domains. Although the workshop did not consider other domains in human welfare, such as disaster management, the themes are likely applicable to those areas as well. The common themes that emerged about the needs to realize the potential for human health applications include the following: Need for integration of spatial data on environmental conditions derived from land remote sensing with socioeconomic data; Need for communication between remote sensing scientists and decision makers to determine effective use of land remote sensing for human welfare issues; and Need for acquisition, archiving, and access to long-term data—both historical and future—and for development of the capacity to interpret data.
OCR for page 37
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Integration of Land Remote Sensing and Socioeconomic Data Remote sensing has been applied to many areas such as weather forecasting, global change research, and local applications (e.g., precision agriculture). The potential for a major contribution to human welfare is largely unrealized. The success of this use of remote sensing rests with the ability to integrate remote sensing data with socioeconomic information. The research community is only at the beginning steps of achieving this integration. There is great value in combining spatial and socioeconomic data for use by policy makers. Additional research on how this integration should occur might be considered. Socioeconomic information such as land management and land tenure is crucial at global, regional, and parcel scales, and the dynamics and differences in patterns and processes of land use that occur between these scales have to be better understood. Data that provide a means of discriminating among different agricultural or land practices (e.g., tilled versus untilled soil, differences in tillage practices) can be valuable information for agricultural policy making. An understanding of coupled urban-agricultural systems will be increasingly important as urban areas continue their rapid expansion and food and amenities are cycled between the two areas. In addition to the benefits of combining socioeconomic and remote sensing at various scales, remote sensing data and information can be used to improve agricultural production on a seasonal basis by identifying where the crop yields are high and low and where drought or other types of water stress are likely to affect certain crops or regions unfavorably. If adaptive management techniques are to be employed, ongoing agricultural monitoring, via remote sensing as well as other methods, is necessary. The data obtained through monitoring can be used to make informed decisions about what areas are likely to experience declines in yields in a specific season and how commodity pricing may be impacted. On a larger scale, remote sensing is currently used to predict El Niño phenomena, which can have significant implications for agriculture. In the human health domain, the integration of socioeconomic information, such as locations and vulnerabilities of human populations and access to health infrastructure, with environmental conditions, such as habitats for disease vectors and potential disease outbreaks, is key to providing information that is effective in generating response strategies. Effective Communication Between Land Remote Sensing Community and Decision Makers To be most useful in decision making, technicians and policy analysts trained in remote sensing and geographic information analysis could
OCR for page 38
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report become part of interdisciplinary teams. Information from such teams could provide useful information that feeds into decisions, for example, to promote sustainability while maximizing crop yields and/or avoiding food or livelihood crises. Particular emphasis could be placed on increasing user capacity within lesser developed countries and building a cohort of individuals with adequate technical and analytical training to take advantage of access to remote sensing data and information. In all parts of the world, communication between technical experts and policy makers is inadequate. Policy makers could become better informed about the utility of remote sensing data, to ensure both that resources are available to utilize these data and that data are fed into all appropriate stages of the decision-making process. Minimum data needs and levels of accuracy would be communicated. In addition, public and private sector data users would understand and implement integrated communications and knowledge management strategies. When remote sensing data and information are disseminated, activities such as culturally specific outreach and data presentation could occur to make the data meaningful to recipients. Human health is the outcome of numerous physical and socioeconomic factors. Because remote sensing techniques can monitor human health only through indirect means, it is important for decision makers and remote sensing scientists to communicate early and often regarding data and knowledge needs and transfer. If a population is experiencing a state of relatively good health and prosperity due to the combination of environmental and socioeconomic factors, it is important to observe any changes in those factors that could adversely affect the population. It is important that the observations are made rapidly in order to employ adaptive management practices. For example, the ability to monitor change could allow decision makers to predict the migration and rate of disease transmission to, from, or within populated areas and improve the ability for health service response during times of crisis or emergency. Decision makers and remote sensing scientists might consider developing a series of short- and long-term priorities in order to determine data needs and develop the ability to implement decision-making strategies. Short-term (five-year) priorities include the following: Understanding ecological processes and how they interact with disease occurrences; Fusing settlement and population data with other types of remotely sensed data into geographic information systems (GIS) layers; Developing risk-based decision-making practices to enable action to be taken even when data do not provide complete certainty;
OCR for page 39
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Providing opportunities for interactions among scientists from different disciplines to understand the ecological and social dimensions of health; and Enabling access to data at a reasonable cost. Long-term (10-year) goals include An end-to-end system to support health applications, including the collection, analysis, and application of data in decision making; and Multi-scale, multi-temporal data for surveillance, monitoring, and prediction. Data Access and Technical Capacity To expand the use of remote sensing data in human welfare decision making, these data have to be more readily available, more affordable, and deliverable in useful formats. Decision makers responsible for allocating resources for remote sensing technologies might consider communicating with remote sensing practitioners to determine what data types and data scales are necessary (i.e., countrywide versus crop specific). Data could be delivered in formats more compatible with GIS applications commonly in use in order to be combined easily with other types of data. Training and capacity building in the use of remotely sensed data could be increased, especially in developing countries. Organizations such as the National Geospatial-Intelligence Agency (NGA), in partnership with the U.S. Department of State, the military, or universities, might consider distributing data for no or low cost in large quantities that could then be made accessible to new practitioners with the ability to deal with large data sets. With the exception of the Landsat program, remote sensing platforms were generally not designed to generate data for current research and applications addressing the links between land cover, environmental conditions, and human welfare. Nevertheless, the data are potentially useful for early warning systems similar to FEWS NET. FEWS NET, now established for many years, may be supplemented by future early warning systems such as a global disease monitoring and early warning system, a poverty monitoring and early warning system, and a biodiversity monitoring and early warning system. These systems would be driven by timely and accurate remotely sensed data. The main challenges in establishing these types of early warning systems lie in the following:
OCR for page 40
Contributions of Land Remote Sensing for Decisions about Food Security and Human Health: Workshop Report Constructing an archive of global environmental data accessible to all; Developing appropriate image-processing algorithms that produce relevant, processed data layers; Developing robust predictive models for biodiversity, agriculture, health, poverty, and environmental changes through time; Linking model outputs to the formulation of environmentally sound policies that are effective at the grass-roots level; and Producing a monitoring and feedback system that returns quality field data from project areas, to improve the modelling process. Since human welfare issues involve understanding change over time, archival data (historical maps, aerial photography, pre-satellite era data, and historical satellite data) are key resources that must be preserved. Continued long-term monitoring and archiving of data from current satellite platforms are vital for the same reasons. It would be beneficial for funding considerations and planning of satellite missions to be made with the long term in mind, although funding streams are often not continuous and current project durations tend to be for only one to two years. Workshop participants identified the lack of time series and real-time data from the National Aeronautics and Space Administration (NASA) and other data sources as major barriers to the application of remote sensing technologies to human welfare improvement. Decisions will also have to be made on whether resources are better spent on the collection of more remotely sensed data or on more data analysis. Potential benefits from the use of remote sensing data and information in food security will be enhanced and made more affordable both by improvements in current technologies and by additional international competition in the development of satellites and sensors, which will increase data availability. Non-satellite technologies for remote sensing could also be advanced, including enhanced aerial photography and the use of unmanned aerial vehicles. Finally, new types of data collection, analysis, and access technologies could be developed to make comparable data for global and regional estimates of the extent of tillage, land use, and land cover change. Because of the growth of urban populations throughout the world and the limited amount of land available for agriculture, agricultural land is a resource that will require careful monitoring and protection.
Representative terms from entire chapter: