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 78
Integrating Multiscale Observations of U.S. Waters 4 Case Studies on Integrated Observatories for Hydrological and Related Sciences Because the importance of hydrological observatories has been recognized for decades, many existing field study sites are available that illustrate how innovative sensor technologies and modeling approaches are being, or could be, applied. In this chapter, case studies drawn from the expertise of committee members and their many collaborators are presented. They represent a broad range of different types of projects in terms of motivation, location, design, and duration of study. Some primarily describe on-going activities while others represent proposed activities that would be overlain on existing but limited monitoring or science programs. Each example is presented to highlight one or more important issues related to the design and operation of hydrologic observatories, test beds, and campaigns. However, this report is not recommending that these specific case studies be undertaken. Further, the individual case studies are intended only to illustrate the kinds of sensors, sensor networks, or data analysis that could be valuable in integrated observation systems, rather than to provide specific advice to any governmental entity. Generalized findings and conclusions derived, in part, from the case studies taken as a whole are found in Chapter 5. INTRODUCTION TO THE CASE STUDIES “Monitoring the Hydrology of the Everglades in South Florida” provides an excellent example of a large, complex, integrated observatory designed for pressing water management needs in an ecologically sensitive area. This case study describes how the South Florida Water Management District (SFWMD) together with the U.S. Army Corps of Engineers and numerous other state, local,
OCR for page 79
Integrating Multiscale Observations of U.S. Waters and tribal partners are working together on issues of water quantity and quality, flood management, and ecosystem protection. The case study describes a large monitoring network of hydrological, meteorological, and other sensors along with the hardware and software infrastructure needed for data collection and management. Some new sensor systems that potentially could be employed to enhance ecosystem monitoring are identified. The case study illustrates the importance of interagency cooperation, which can be crucial to the success of complex observatories for hydrological and related sciences. “Impacts of Agriculture on Water Resources: Tradeoffs between Water Quantity and Quality in the Southern High Plains” addresses the impacts of agriculture on water resources, with a focus on semiarid regions where water availability is a critical issue and where cycling of salts has large-scale impacts on water quality. This type of study is particularly important given that world food needs will continue to increase, that many nations are turning to biofuels, and that climate change may worsen drought in many parts of the world. The case study discusses the types of measurement and monitoring programs that should be conducted to provide the necessary information to develop sustainable water and land resource management programs in the High Plains. In the High Plains, large time lags exist between forcing (land-use change) and response (increased recharge; change in water quality). Hence, the study provides a ‘classic’ example of why long-term observations often are critical, and why it is important that observatories have the capability to endure through changing budget cycles. “Hydrological Observations Networks for Multidisciplinary Analysis: Water and Malaria in Sub-Saharan Africa” extends the case study examples to include the Developing World and direct issues of world health and medicine. This study demonstrates the importance of establishing consistency in sampling locations for different parameters, in this case analysis of climate and hydrological conditions versus malaria outbreaks. Proper coordination of physical, chemical, biological, and medical data collection, at appropriate spatial and temporal extents, is a key to inferring the controls on malarial outbreaks or the best methods for preventing such outbreaks. This case study thus emphasizes the axiom that the nature of the research question or research hypothesis plays an important role in the design of the associated observation network. “Achieving Predictive Capabilities in Arctic Land-Surface Hydrology” explores a rudimentary strategy for robust remote sensing hydrology in the PanArctic, to identify capabilities needed to link in-situ observations to satellite sensor-scale observations. The assertion in this case study is that the appropriate mechanism for achieving this linkage is through robust models that span the scales of the hydrologic processes. Given the difficulty of access to Arctic sites and the sensitivity of the Arctic ecosystems, there is a pressing need for autonomous sensing stations at point- through plot-scales, airborne platforms for plot-through watershed scales, and satellite remote sensing for sub-watershed through Pan-Arctic scales. This case study therefore highlights the need for collecting
OCR for page 80
Integrating Multiscale Observations of U.S. Waters and integrating data at different scales and for developing novel methods for working in remote and challenging environments. “Integrating Hydroclimate Variability and Water Quality in the Neuse River (North Carolina, USA) Basin and Estuary” focuses on the impact of human activity and hydroclimate variability on watershed nitrogen sources, cycling and export, and their impacts on fresh water and estuarine ecosystem health. As outlined in this case study, the problem requires a synthesis of hydrologic, ecosystem and anthropogenic water, carbon, and nutrient processes within a coupled watershed and receiving estuary. Nutrient management in the Neuse watershed focuses on nitrogen, which is the limiting nutrient in the estuarine system. New sensing methods are emerging that promise to transform our ability to measure and understand nitrogen ecosystem dynamics. Hence, this case study demonstrates the need for integrating measurement of hydrologic, biogeochemical, and other ecosystem-related processes, and for building coordinated teams with interdisciplinary capabilities. “Mountain Hydrology in the Western United States” explains how snow in mountains of the West is the main source of the region’s water, with downstream hydrologic processes (e.g., groundwater recharge) and interactions with ecosystems controlled by processes at higher elevations. Hence, it is critical to develop models for water and energy fluxes in the western mountains that can take into consideration not only past and present conditions but likely changes brought about by climate change. The critical issue is the need for high spatiotemporal resolution due to sharp wet-dry seasonal transitions; complex topographic and landscape patterns; steep gradients in temperature and precipitation with elevation; and high interannual variability. Given the need for high spatiotemporal resolution data, this study illustrates the need for developing and taking advantage of emerging embedded sensor network technologies, coupled with already existing monitoring and modeling strategies.
OCR for page 81
Integrating Multiscale Observations of U.S. Waters CASE STUDY I —MONITORING THE HYDROLOGY OF THE EVERGLADES IN SOUTH FLORIDA The Everglades The Florida Everglades (Figure 4-1) is one of the world’s largest freshwater wetlands. It was once a free-flowing river of grass that provided clean water from Lake Okeechobee to Florida Bay. The marshes and swamps acted as natural filters that recharged underground aquifers in the South Florida region. Historically the pre-channelized Everglades hydrologic balance was maintained through long, slow, continuous, gravity flow of water. Because of the diversion of water, channelization of transient rivers, and loss of elevation through oxidation of soils, pump stations are now required to move water from canals to marsh areas or from one canal segment to another or to return seepage water that would otherwise be lost from the greater Everglades. Over 50 such stations now exist, pumping volumes ranging from ~200 cfs to ~4800 cfs (Susan Sylvester, SFWMD, written commun., November 2006). Accordingly, today the releases from Lake Okeechobee are controlled. During normal climatic conditions, Lake Okeechobee outflows are able to meet the large water needs to the south of the lake. However, when the climate remains abnormally dry for an extended period (for one or two seasons), inflows may diminish to very low levels during the same period that demands on the lake will peak. Consequently, lake stages may fall very quickly to extremely low levels. Conversely, when climatic conditions are wetter than normal, large volumes of water enter the lake, coinciding with periods when water demands to the south will be minimal. These events cause lake stages to rise very quickly and require large volumes of water to be discharged to the Water Conservation Areas (WCAs) or to the St. Lucie and Caloosahatchee estuaries. Abrupt changes in flow or very large releases through the estuaries are harmful to these ecosystems. The WCAs are the primary source of supplemental water for the highly developed urban areas along the southeast coast of Florida, with the lake being the alternate source. The WCAs were built as large water-storage impoundments in the Everglades to provide both water supply and flood protection for the urban areas. In addition to the agricultural and municipal water consumptive needs, water releases from the lake are required to meet the needs of the Everglades and the numerous coastal ecosystems. The WCAs and the Everglades National Park (ENP) are known today as the remnant Everglades. Water held in and released from the WCAs effectively recharges the Biscayne aquifer in some areas. Over the past half-century measures taken to satisfy agricultural and urban development goals have degraded the Everglades ecosystems. To restore and maintain the vitality of these ecosystems as well as to enhance the reliability and quantity and quality of water supplies, and provide flood protection, the U.S. Army Corps of Engineers (USACE), the South Florida Water Management District and numerous other federal, state, local, and tribal partners involved in water
OCR for page 82
Integrating Multiscale Observations of U.S. Waters FIGURE 4-1 Map of the Everglades region in South Florida. The South Florida Water Management District is responsible for managing the hydrology and ecology in this area. SOURCE: NRC (2006a). © International Mapping Associates.
OCR for page 83
Integrating Multiscale Observations of U.S. Waters management in South Florida, have developed a plan called the Comprehensive Everglades Restoration Plan (CERP). To learn how to better manage the water, and to better understand the impact of various regimes of water-quantity flows, stages, volumes and qualities, and their durations and timing, timely, comprehensive, and accurate monitoring information is essential. Considerable sums of money have been spent in establishing an elaborate hydrologic, meteorological, and water-quality monitoring system throughout the Everglades. This is an excellent example of how an integrated hydrological observatory can provide essential data for managing water quality and quantity, flood control, and ecosystem protection. The South Florida Water Management District The South Florida Water Management District (SFWMD or the District) is the primary agency responsible for monitoring, managing, and protecting water resources in a 46,439 km (17,930 mi) region of South Florida. The District operates approximately 3000 km (1800 mi) of canals and more than 200 primary water control structures to serve a population of over 7 million people. The District’s annual budget exceeds $1 billion of which some $20 million (about 2 percent) is spent on hydrologic monitoring and associated data management activities. (This number would be larger if such activities in the area of water quality were included.) The hydrologic monitoring network of the District is divided into five parts: (1) rainfall, (2) meteorological, (3) surface-water stage, (4) surface-water flow, and (5) groundwater. These networks are spatially distributed over the geographic areas of the District. For each network, the District maintains records on the history and evolution of the network; information on sensor(s)/instrument(s) used; number and location of instruments; frequency of data collection; time interval of the available data; optimization or design of the network conducted; and relevant references used. The District has been gathering data about the region’s water and land resources for more than 40 years. Information about past and current weather, rainfall, and changes in vegetation or land use is essential for current and future planning, operations, research, and restoration initiatives. Real-time data, especially when combined with historic data, help the District make more informed water resources management decisions. Information about how natural and manmade systems are working (or not)—individually and interactively—is essential to short and long-term water resources management and restoration. Data Collection and Management Modern electronic hydrologic monitoring, data collection, and management
OCR for page 84
Integrating Multiscale Observations of U.S. Waters began at the District in 1974. This has allowed water managers to remotely monitor strategic flood gates and control hydrologic conditions. The backbone of this system today is a 24-station microwave infrastructure with two-way radio extensions. This recently modernized microwave communications infrastructure now supports voice radio relay, supervisory control and data acquisition (SCADA), telephone circuits/trunks, and computer network traffic. SCADA systems include hardware and software components that scan all remote data, log data and system events, send alarms when abnormal conditions occur, and issue operator commands to remote devices. The District’s SCADA and Hydro Data Management (SHDM) Department is responsible for data collection and management (Figure 4-2). The District’s SCADA system transmits and receives information on water stages or levels, wind velocities, rainfall, water temperature, salinity levels, and other data. The system operates continuously and uses wireless communications to monitor and control water level, water control gate positions, and pumping activities. It provides an early warning of possible flood problems by observing water level and rainfall trends. This computerized data collection system comprises the cornerstone of the District’s data collection through a District-wide network of real-time and near-time data collection stations. The District also obtains and processes a variety of manual data logs. Hydrologic data management includes processing the data collected, summarizing, deriving and analyzing, storing, and publishing. Processed data are archived into two different databases, namely, Data Collection/Validation PreProcessing (DCVP) and DBHYDRO. Instantaneous (breakpoint) data are stored in the DCVP database, while daily summary and 15-minute interval data are published in the DBHYDRO database. End users can retrieve data from either of these two databases. DBHYDRO data are accessible to users through the web browser. Internal users can also retrieve information from the DCVP archive using the web browser. The District maintains a structured quality assurance/quality control procedure to ensure that data collected is of the best possible quality before it is further published. Pre-processing is the first stage of operations applied to “raw” time series data collected within the District’s monitoring network. Data records are collected and posted to DBHYDRO after data processing. Data quality assurance is normally performed during data processing. However, for some select legally mandated sites and for baseline data used in regional modeling and CERP, some post-processing quality assurance/quality control including graphical plotting and statistical analysis, are also performed. Monitoring Networks As of April 30, 2007, the District actively operates and maintains a network of 287 rain gauges to obtain rainfall data. These data are supplemented by radar
OCR for page 85
Integrating Multiscale Observations of U.S. Waters FIGURE 4-2 The hydrometeorological data flow and process. SOURCE: Pathak (2008). rainfall NEXRAD (Next Generation Radar) data. The District also operates and maintains 45 active weather stations. In addition, data used to estimate daily potential evapotranspiration (PET) by “the Simple Method” several methods are available for 19 weather stations. A network of 1265 active surface-water stage gages provides surface-water stage data for various water bodies. Additionally, the District owns and operates a network of 446 active surface-water flow monitoring sites that provide instantaneous flow data at 15-minute intervals. From these data mean daily flows data are derived. The groundwater monitoring network has a total of 905 groundwater wells that are monitored on an interval basis of 15-minute, monthly, or greater than 1 month. The hydrologic monitoring network at the District is dynamic in nature and is constantly being expanded and optimized to the changing needs of the District. Rainfall Measurements: The Importance of Employing Complementary Methods An example of how different methods of data collection can complement one another is shown by the NEXRAD system. NEXRAD or Weather Surveillance Radar data provides complete spatial coverage of rainfall amounts unob-
OCR for page 86
Integrating Multiscale Observations of U.S. Waters trusively using a predetermined grid resolution (usually 2 km × 2 km or 4 km × 4 km). The NEXRAD rainfall data is limited by reliance on the measurement of raindrop reflectivity, which can be affected by factors such as raindrop size and signal reflection by other objects. Because the reflected signal measured by the radar is proportional to the sum of the sixth power of the diameter of the raindrops in a given volume of atmosphere, small changes in the size of raindrops can have a dramatic effect on the radar’s estimate of the rainfall. For this reason, the radar is generally scaled to match volume measured at the rain gages. The best of both measurement techniques is realized by using rain gage data to adjust NEXRAD values. Surface-Water Flow Monitoring Network: Linking Monitoring with Control The District operates a network of 446 active flow monitoring sites that are used in operations, planning, and regulatory aspects of water management. The flow monitoring network is shown in Figure 4-3. The District works closely with the U.S. Geological Survey (USGS), USACE, and various local agencies in measuring and/or estimating flow throughout the District’s water control facilities. Water control structures are used to divert, restrict, stop, or otherwise manage the flow of water. These water control structures include pump stations, spillways, weirs, and culverts. District structures are typically designed to operate under a combination of water levels and operating regimes, which in turn result in varying flow conditions. Flow that moves through the structures are estimated by using a rating equation appropriate for the flow conditions based on the structure’s static and dynamic data. The “static” data include the geometric characteristics of the structure, whereas the “dynamic” data comprise the water stages (headwater and tail-water) and operating conditions (gate opening and pump speed). Groundwater Monitoring Network: An Example of Interagency Cooperation Groundwater monitoring data are needed to assess long-term trends in groundwater availability; to develop, verify, and calibrate groundwater flow models; to assess temporal groundwater conditions during droughts; to provide data for water-use permit application evaluations; to assist the District in legal proceedings involving regulatory and other groundwater disputes; and to use in designing and evaluating various District projects. The District groundwater network consists of wells that have data publicly available through the District’s DBHYDRO database but also in other databases not publicly accessible (these are mostly project specific). There are ongoing plans to migrate both of these databases into the DBHYDRO database. The groundwater network also con-
OCR for page 87
Integrating Multiscale Observations of U.S. Waters FIGURE 4-3 District flow monitoring network. SOURCE: Reprinted, with permission, from Pathak (2008).
OCR for page 88
Integrating Multiscale Observations of U.S. Waters sists of wells monitored by USGS through a cooperative agreement with the District. Most of the data on these wells are also available in DBHYDRO, but some can only be assessed from USGS’s Automated Data Processing Systems (ADAPS) database. The District measures groundwater levels by using a pressure transducer, typically connected to a data logger. Thepressure transducers measure head pressure. The transducers communicate with the data loggers through an electronic cable. The data loggers then convert measured pressure values into water levels and record these data for subsequent downloads via laptop computers. Alternatively, data from some of the wells connected to the data loggers are sent via telemetry. Ecological Monitoring: Merging Established and Emerging Approaches The Everglades/Florida Bay landscape is a mosaic of different habitats that have evolved under a highly dynamic set of environmental conditions. As with any complex system, interactions among its different components are a fundamental aspect of its operation and play an important role in sustaining the Everglades. The physical hydrology, biogeochemical nutrient cycling, and biology of plant and animal communities are determinants of the emergent ecosystem properties that comprise the landscape. Monitoring these different “processes” that “drive” the system is providing data on how to best restore and maintain this dynamic landscape. Monitoring the complex hydrologic, floral, and faunal changes associated with restoration activities is an enormous task. Many physical, chemical, and biological parameters have been identified as measures, or indicators, of overall performance of proposed restoration activities. Carefully designed and methodically implemented system-wide monitoring strategies are needed to successfully quantify both short- and long-term changes within the interdependent landforms, vegetation assemblages, and animal communities. Due to the vastness (some 3 million acres) of the Greater Everglades, including the surrounding agricultural and urban environments, fixed monitoring stations (e.g., stage gages, water-quality collection sites), and vegetation/soil field sampling schemes (e.g., points, transects) cannot yield the high density of sampling data needed to adequately characterize and model the diverse ecosystems. Remotely sensed data, which is able to cover large areas with uniformly distributed high-density data points, has been and will continue to provide the essential synoptic view of restoration activities and their effects. Work is underway to develop realistic and attainable strategies to expand the effective utilization of remotely sensed data for CERP system-wide adaptive monitoring and assessment. This involves the participation of biologists, chemists, hydrologists, and engineers in matching potential remote sensing technologies (sensors and analysis techniques) with the data required for effective water
OCR for page 132
Integrating Multiscale Observations of U.S. Waters FIGURE 4-18 Persistence of snow-covered area in the Colorado River Basin, western United States over the 1995-2002 period for March and April, reported as number of years each 1 km2 Advanced Very High Resolution Radiometer (AVHRR) pixel had detectable snow cover during that month. Snow pack, which covers a relatively small fraction of the basin, provides over 85 percent of the annual discharge in the Colorado River. SOURCE: Reprinted, with permission, from Bales et al. (2008). © 2008 by the American Geophysical Union. While research challenges span water issues in the West, the potential for integrated observation systems involving new measurement technologies to impact these issues is perhaps greatest in mountain hydrology (Bales et al., 2006). In the mountains of the western United States, sharp wet-dry seasonal transitions, complex topographic and landscape patterns, steep gradients in temperature and precipitation with elevation, and high interannual variability make hydrologic processes and variations significantly different from lower-elevation regions or those that are humid all year. Hydrologic feedbacks in mountainous regions control the availability of water, influence the distribution of vegetation, dominate biogeochemical fluxes, and contribute to global and regional climate variability. Snow in mountains of the West is the main source of the region’s water, with downstream hydrologic processes (e.g., groundwater recharge) and interactions with ecosystems controlled by processes at higher elevations.
OCR for page 133
Integrating Multiscale Observations of U.S. Waters Despite the importance of mountain regions to the hydrologic cycle, the processes controlling energy and water fluxes within and out of these systems are not well understood. Further, the lack of integrated measurement strategies and data/information systems for hydrologic data hamper improvements. As examples, we lack a robust framework for accurately describing and predicting the partitioning of snowmelt into runoff versus infiltration and into evapotranspiration versus recharge (Figure 4-19), and we lack strategies to exploit emerging technology to more accurately measure the spatial variability of snow cover and soil moisture in the mountains. The volume of mountain-block and mountain-front recharge to groundwater and how recharge patterns respond to climate variability are poorly known across the mountainous West (Earman et al., 2006; Wilson and Guan, 2004). Three aspects of the mountain water cycle in the western United States are used to illustrate the needs and opportunities: (1) precipitation and microclimate, (2) snow pack, and (3) soil moisture. In each of these, researchers must determine the optimal sampling strategy based on such considerations as cost versus resources, reliability, effectiveness, variability of the parameters in question, sensitivity of models to the given parameter(s), potential for new methods to make a substantial impact, and needs of regulators and managers. One important application for this new knowledge is illustrated, hydrologic forecasting, including the water/energy cycle coupling and data integration needed for the emerging generation of forecast tools. Microclimate and Precipitation Microclimate can vary substantially at a scale of meters in the mountainous West, with large diurnal fluctuations. Direct measurements of precipitation in mountain environments are particularly challenging, because of the need to cover a large range of elevations and orographic positions. Mountains often have too much topographic variability to effectively use the Doppler Radar systems that have proven useful for monitoring extent and intensity of precipitation throughout the Midwest. Moreover, much of the precipitation falls as snow, and precipitation gages catch too little of the snowfall. Although it is possible to infer snowfall rates from snowpack accumulation measurements, direct observations of precipitation are needed for research (for example, to explore changes in precipitation type [e.g., Knowles et al., 2006] associated with climate variability and change), and for applications, such as to drive flood forecasts. The spatial patterns of precipitation in mountainous terrain are nearly impossible to measure at the resolution of basin-scale hydrologic models (e.g., ~1 km). The primary ground-based resources available to estimate these patterns are several existing, overlapping networks, including NWS cooperative stations, RAWS (Remote Automated Weather Stations), USDA SNOTEL (Snowpack Telemetry) stations, and some smaller networks. Most of these stations lack the
OCR for page 134
Integrating Multiscale Observations of U.S. Waters FIGURE 4-19 Schematic of the inter-related fluxes comprising the mountain water cycle, and partitioning of snowmelt. Photograph courtesy of Noah Molotch, University of California, Los Angeles. capacity to differentiate snow from rain. The National Weather Service installed the Next Generation Radar system (NEXRAD) in 1994 to improve operational measurements of precipitation around the country. However, NEXRAD signals are occluded by mountains and thus are less reliable in the complex terrains where snowfall occurs. Even where precipitation is measured in situ, wind effects limit measurement accuracy. Traditional precipitation gages catch too little snow and cannot discriminate solid from liquid precipitation; when they catch snow, the measurement typically registers when the snow caught by the gage melts, not necessarily when it falls, causing a temporal lag.
OCR for page 135
Integrating Multiscale Observations of U.S. Waters Satellites offer an advantage over the operational radars in producing spatially distributed estimates in mountainous areas due to the unobstructed field of view. The high spatial and temporal resolution of precipitation estimates and the short latency of data availability make geostationary satellites the platform of choice for operational applications. However, microwave data from polar orbiting satellites or combinations of data from geostationary and polar orbiting platforms offer more reliable estimates for applications for which coarse resolution and long data latency are not of concern (Anagnostou, 2004). Distinguishing snowfall from rainfall remains a significant problem in satellite precipitation estimation, with current approaches relying on temperature thresholds or on the use of microwave satellite sensors (e.g., the Advanced Microwave Sounding Unit—AMSU). A network for mountain precipitation, along with associated microclimate measurements important for determining energy fluxes and energy/water interactions, would ideally take advantage of both strategically placed ground sensors and satellite-based (or airplane) remote sensing. However, a major limitation is the availability of accurate precipitation sensors for remote deployments. Other components of microclimate should be measured spatially using a combination of a relatively few well-instrumented conventional measurement stations that offer low spatial resolution but multiple types of well-calibrated and conventionally accepted measurements, complemented by widely distributed embedded sensing devices that greatly increase spatial sampling resolution, but with cheaper probes and fewer types of measurements. For example, temperature and solar radiation (at a variety of wavelengths) probes can be attached cheaply to a microprocessor with radio transceiver and solar battery for wide distribution in a sensor network. For integrated, comprehensive, water balance measurements, at least some of the sensor nodes should be equipped with additional measurement devices such as wind speed and direction measurement; atmospheric moisture sensors, snow, soil moisture and temperature and other sensors. Installing these sensors in an embedded sensor network would help to control more complex sensing devices, to “flag” events in real-time for greater attention by researchers or managers, and to interact directly with models and control devices. Use of digital cameras to visualize the snowline and weather conditions, and potentially to read gages, is feasible provided icing can be controlled. Scientists must often assess the trade-off between more expensive but potentially more labor-intensive measurements of a wide variety of parameters versus highly distributed inexpensive sensors for fewer parameters, some of which may be at lesser precision and accuracy. Given the physiographic variability of mountains, network design that encompasses both types of sampling strategies, along with remotely sensed information, is likely to result in significant success (Figure 4-20). It will be important to use remotely sensed data in the initial sampling network design, and as a potential independent method of data validation and verification. On the other hand, the sensing network should help to ground-truth remotely sensed information. Together with modeling, the combination of methods will provide a strategy for merging data at different scales, hence maximizing spatial and temporal resolution. Models can also help guide measurement design.
OCR for page 136
Integrating Multiscale Observations of U.S. Waters FIGURE 4-20 Conceptual design and deployment of instrument clusters in a mountain basin, integrating satellite remote sensing with strategically placed ground measurements. Selected instrument clusters are anchored by an eddy-correlation flux tower extending above the forest canopy, with ground measurements extending 1-2 km from the tower. Other clusters would consist of sensors and sensor networks but not a tall tower. Adapted, with permission, from Conklin et al. (2006). © 2006 by the American Geophysical Union. Special consideration will need to be made for engineering embedded sensor pods for harsh and variable mountain conditions, and to designing the sensor network in concert with the topographic challenges (e.g., appropriate relay stations to get around obstacles to wireless communication). In some locations, there may be the need for sensor network data from an array of local sensors to be stored on-site for periodic download rather than telemetered directly to the observer. Snowpack Properties Hydrologic and land-surface models are increasingly including mass balance and physically based snowmelt models (e.g., Cline et al., 1998). While several of such models are being used by the research community, improvements in the treatment of snowmelt as well as other model components depends on data. For snowmelt, data needs include spatially distributed components of the surface energy balance, as well as snow properties (Figure 4-20). Ground observations of snow water equivalent (SWE) have been used in conjunction with remotely sensed snow-covered-area (SCA) data to estimate the spatial distribution of SWE across mountainous watersheds (Fassnacht et al., 2003). However, such efforts cannot be expected to yield representative measures of snow distribution across a basin owing to the non-representative location of SWE measurements (Molotch et al., 2006).
OCR for page 137
Integrating Multiscale Observations of U.S. Waters Several sources of seasonal snowcover data exist, ranging from information collected as part of weather monitoring to hydrologic data from networks dedicated to snow data collection, and more recently to remotely sensed products from polar orbiting and geostationary satellites. Remote sensing is the only practical way to measure the spatial extent and variability of snow cover and albedo, and, during the past decade, methods for mapping snow-covered area from visible and infrared instruments on satellites have become well developed (Dozier and Painter, 2004). However, none of the satellite data sets encompass enough system interactions to be considered “snow system” data (Bales et al., 2006). Snow water equivalent is measured at over 1700 points in the Western United States, from a combination of manual snow surveys and transmitting snow pillows. While this large number of samples provides regional knowledge of the spatial distribution of SWE, it is insufficient to resolve the variability of SWE and snowmelt at the basin scale (Figure 4-21). Moreover, most of snow courses and automated stations are situated on flat or nearly flat terrain, and are preferentially placed high enough to that they stay snow covered most of most winters (to justify the expenditures), which means that we are largely blind at altitudes where snow pack is more ephemeral. Also, there are no stations at the highest elevations, which contribute most of the late-season snowmelt. Glaciers, which also help sustain baseflow after seasonal snow has disappeared, are also severely undersampled. The existing snow measurements are used as indices of streamflow, rather as direct measurements of basin-scale snow volumes. A comprehensive snow-measurement network would blend sparse but detailed, accurate measurements of snow water equivalent, microclimate, and other water-balance variables with satellite remote sensing and spatially extensive lower-cost measurements of snow depth and other low-cost sensors. The ground-based system would ideally be composed of low-cost sensors in embedded sensor networks (Figure 4-22) that build outward from existing ground-based snow pillows and snow courses to capture the physiographic variability across a catchment (Molotch and Bales, 2005, 2006). Building outward from existing measurement sites takes advantage of their long and valuable record of measurement, and provides the additional measurements needed to use distributed depth measurements. As snow density varies much less spatially than does snow depth (Molotch et al., 2005), many more depth than SWE measurements are needed. Low-cost temperature sensors can also be used to infer snow depth, and other technologies are under development. Satellite data can reliably and accurately provide SCA information. Various interpolation and modeling strategies are available to blend the ground-based SWE and satellite SCA to provide basin-scale estimates of SWE for predictive modeling. Periodic SWE measurements from aircraft platforms can also contribute to this mix (e.g., gamma ray sensors). Thus snow properties are an excellent example of how relatively modest investments in integrated observation systems can immediately provide critical data products.
OCR for page 138
Integrating Multiscale Observations of U.S. Waters FIGURE 4-21 Contributions of various 300 m elevation bands to snowmelt in Merced River Basin, Sierra Nevada, California. Data derived snowcover depletion based on MODIS satellite data. Fraction of basin in each elevation band is given for reference. Continuous, ground-based snow measurements are limited to three sites in the basin (2100-2500 m elevation). SOURCE: Reprinted, with permission, from Bales and Rice (2006). © 2006 by the American Geophysical Union. FIGURE 4-22 One pod in a sensor web that was installed to measure snow depth over a 20,000 m2 study area in Yosemite National Park. Depth sensor is on right and wireless pod on left. Photograph courtesy of Margot Wholey Photography.
OCR for page 139
Integrating Multiscale Observations of U.S. Waters As satellite radar retrievals of SWE advance, they will and can be integrated into the spatial interpolation schemes developed around the in-situ network. However, for the foreseeable future, there are no operational radar satellites with sufficient spectral and polarization capabilities to infer snow water equivalent in the mountains. A ground-based observational network remains an important component of the snow observing system that will not be replaced by the satellite system because of inherent uncertainties in retrievals. Satellite-based snow cover/depth observations might cover the future needs of some users of snow data, but economics and scientific objectives now require a merging of all available snow information in an enhanced data set. Judicious and strategic extensification of in-situ measurements to lower and higher elevation sites coupled with advancements in remote sensing acquisitions will provide the means to a long-term, high-resolution monitoring of snow pack properties. Soil Moisture Soil moisture is a primary state variable of the land surface. In mountains, soil moisture is greatest following spring snowmelt and lowest in late summer and fall after the soil dries. Its spatiotemporal variability affects surface and subsurface runoff, modulates evaporation and transpiration, determines the extent of groundwater recharge, and initiates or sustains land surface-atmosphere feedbacks. Soil moisture is influenced by: (1) precipitation history, (2) soil texture, which determines water-holding capacity, (3) land-surface slope, which affects runoff and infiltration, and (4) vegetation, land cover, and bedrock slope/depth, which influence evapotranspiration and deep percolation. The partitioning of soil moisture to groundwater recharge, evapotranspiration, and surface/subsurface runoff at different spatiotemporal scales and under different hydroclimatic conditions poses one of the dominant challenges in quantifying water cycle variability (Jacobs et al., 2006). Soil moisture is measured at some RAWS and SNOTEL sites, but these generally do not lie in mountain settings. Remote sensing of soil moisture is made in the microwave frequencies. As in the case of SWE, passive microwave retrievals of soil moisture are too coarse for the spatial variability and rugged terrain in mountain regions. Radar retrievals of soil moisture (and SWE) are in research mode now but again the lack of an operational radar satellite presents a significant obstacle to the implementation of these retrievals (Shi et al., 2002; Oldak et al., 2003; Western et al., 2004). Moreover, rugged terrain and vegetation cover can confound retrievals. While the technical difficulties are great, definition of the spatial variability of soil moisture is critical in modeling hydrologic response in mountain catchments (Zehe and Blöschl, 2004). The strategy for measuring soil moisture in mountain catchments mirrors that for snow depth, that is, design a network that captures the spatial variability
OCR for page 140
Integrating Multiscale Observations of U.S. Waters in physiographic features (e.g., north- versus south-facing), plus distance from trees. A number of low- to moderate-cost soil moisture sensors are available that can be incorporated into sensor networks, and new technology is emerging. As radar retrievals of soil moisture advance, they should be integrated into the spatial interpolation schemes developed around the in-situ network. Like SWE, proper network design for soil moisture measurement remains a research issue. Data Integration and Applications An additional, more-general challenge that cuts across all aspects of water in the West is that of data integration. Current computing environments, investigator-specific research practices, and agency data distributions are disjoint and do not readily facilitate system/data integration. Typically, these systems use ad hoc scripts to perform the required processing and idiosyncratic naming conventions for the files that hold the products. Data extraction from a variety of systems is, therefore, time-consuming and subject to error proliferation, especially when assembling a synoptic view or parsing the data according to a suite of criteria (e.g., data from all snow courses above some particular elevation in a particular basin with more than some threshold length of record). Our current modes of analysis usually require reorganization of data and creation or rediscovery of metadata values for each product. Dissemination, especially where custom processing such as subsetting, reprojection, or reformatting is required, is often treated in a similarly ad hoc fashion. The technologies to solve most of these problems are at hand already, but implementation will require a concerted, collective commitment by users and providers (Bales et al., 2006). Cyberinfrastructure advances can overcome many of the current data problems by making data and information available in ways that are convenient for users. That does not necessarily mean that data are made available to users in the same way that they access data now. Rather, technologic advances that are tailored to be responsive to community needs can both make users more cyber-savvy and information more accessible. Hydrologic forecasting is used to illustrate the role of new, integrated measurement systems. Hydrologic forecasting methods for operational management are well established, and are based on several decades of historical data from sparse networks that monitor surface precipitation and temperature, snow pack, and river stage or discharge. Specifically, flow-forecast reliability and accuracy depend critically upon the use of historical data to calibrate the operational hydrologic models for the watersheds of interest (Fread et al., 1995; Finnerty et al., 1997) and upon the use of reliable quantitative precipitation forecasts (Olson et al., 1995; Sokol, 2003). However, forecast skill for spatially distributed flow forecasts over small areas in the populous coastal mountainous basins or for seasonal forecasts of reservoir inflows in large reservoir projects on the Sierra Nevada is in many cases poor for effective emergency and water supply management.
OCR for page 141
Integrating Multiscale Observations of U.S. Waters Long-range reservoir inflow forecasts depend largely on long-range forecasts of surface climatic variables (temperature and precipitation). Nevertheless, reliable estimation of forecast uncertainty for such long-range forecasts will benefit directly from improved good quality hydrologic observations, and recent integrated forecast-management demonstration projects show that reliable forecast uncertainty is necessary for improved reservoir management (Yao and Georgakakos, 2001; USACE, 2002; Georgakakos et al., 2005). The improvement of flow forecasts over small areas will require improvements in high-resolution measurement technology for precipitation and temperature (NRC, 2005) and the use of real-time flow-forecast assimilation procedures (Seo et al., 2003). Here again, good quality data will contribute directly to improved emergency management effectiveness. In addition, future conditions (e.g., warmer temperatures, snow-to-rain transitions) are expected to be outside the range of past system behavior. Several recent studies find evidence that climatic impacts on western water resources are in transition to a new regime (Stewart et al., 2005). This would make large improvements in hydrologic observations and information even more important. Summary Mountain snow pack is the main source of the American West’s water resource. However, this resource is particularly vulnerable to ongoing spatial and temporal changes in melt patterns, which will directly affect the seasonal availability of water to the multitude of stakeholders in the region. Three aspects of the mountain water cycle in the western United States were used to illustrate the region’s needs and opportunities: (1) precipitation and microclimate, (2) snowpack, and (3) soil moisture. The integration of many types of observations can assist the management of mountainous water resources. In most cases, the resulting data and information need to be continuous, reliable, and rapidly available for effective prediction and management. For these reasons, effective strategies such as the use of embedded sensor network technology would contribute greatly to the progressive use of new data sources and types in the water resources management process and for quantifying their benefits and associated costs. The organization and careful monitoring of prototype demonstration projects with the participation of measurement specialists, scientists (including communication and computer scientists), forecasters, and managers is an effective means to develop and test such strategies. To summarize the chapter, the six case studies presented above are located in climatic regions from subtropical to semiarid to alpine highland to circum polar. They range from large scale to medium scale, and from programs that are largely ongoing to initiatives that are nascent or proposed. Some are more oriented toward scientific understanding; others are oriented toward improving man-
OCR for page 142
Integrating Multiscale Observations of U.S. Waters agement decisions. Because of this, they provide a wide variety of lessons for government agencies, academic institutions, and even the private sector. These lessons are summarized and discussed in the following chapter.