Southern California faces many challenges in managing our environment, as do other regions of the country. How do our human activities affect, and in turn are affected by, changes in ecosystem structure, regional climate, and land use? How will future changes in land use and predicted changes in climate in Southern California influence fundamental ecosystem processes and critical services on which we depend so strongly? What will be the spatial and temporal scales of such changes and will responses be gradual or abrupt? At their roots, these questions reflect not just a curiosity about our future, but a desire for proactive management to mitigate undesirable outcomes. We would like to know the magnitude, pace, and geography of ecological changes, and to understand the implications of such changes for our environment and the provision of ecosystem services to humans. Answers to these questions will require new technologies in environmental monitoring and remote sensing.
In past years, articles in the Southern California Environmental Report Card have dealt with a variety of regional issues, for example air pollution, groundwater pollution, biodiversity, and invasive species. While it is appropriate to address these important issues with focused attention, we also need to understand the complexity of our environment and the interaction between individual drivers of change. Human populations are initiators of ecosystem change as well as responders to such changes. Growing human population drives increases in urbanization and suburbanization in Southern California, and our needs for housing, commerce, food, and recreation have consequences for the structure and function of both natural ecosystems and our human environment. Changes in land use and land cover give rise to fundamental changes in ecological structures and processes, and thus to the ecosystem services that feed back to us. We all benefit from, and indeed depend on, a relatively high level of sustainability in water resources, soil fertility, climate conditions, and biodiversity.
Resource managers and policy administrators in the past have generally perceived ecosystems, and the organisms they contain, as passive responders to climate variation and climate change, with no reciprocal effect on weather and climate. We now know, however, that ecological processes in one part of a region may have impacts on distant areas through the atmosphere and its circulation, and through flows of water through drainage networks. The realization that land use and ecosystem structure have important feedbacks to weather and climate has been a transforming paradigm for sustainable management of our environment.
Appropriately addressing all of these aspects of environmental impacts and interdependencies will require an entirely new generation of technology innovations. Fortunately, three important areas of technology advancement are occurring at the present time: new technologies for sensors and sensor platforms, remote sensing technologies, and technologies for efficient data transmission and analysis. This article describes these important new technological developments, and highlights examples of their applications to key environmental problems.
The rapid development and miniaturization of technologies used in digital cameras, cell phones, and wireless computers are allowing scientists to develop networks of small sensors that will lead to a new era of monitoring the health and stability of our environment. Wireless devices half the size of a cell phone now exist with sensors to measure light, wind speed, rainfall, temperature, humidity, and barometric pressure. Moreover, these devices store collected data, process desired data averages or transformations, and then transmit requested data by radio frequency along a series of wireless hops to an Internet node.
Deploying arrays of hundreds of these sensor devices will allow us to fill a gap between local-scale ecological observations and environmental data from scattered regional weather stations. Such micrometeorological measurements at fine spatial and temporal scales will help scientists understand the relationship of broad-scale changes in global climate and local microclimate that control many ecosystem and physiological processes. These sensor networks have the potential to revolutionize science and to influence major economic, agricultural, environmental, social, and health issues, as well as to enhance opportunities for new educational programs.
Beyond fixed arrays of meteorological sensors, new types of sensors and improved sensor platforms are also being developed to provide environmental scientists with significant tools for understanding fundamental ecosystem processes in a manner not previously possible. Multi-spectral video imagers, acoustic sensors, gas analyzers, and other high-performance instruments are now being added to remote, unattended network deployments (Figures 1 and 2) . These technologies greatly expand our ability to monitor the environment to understand patterns of global change and changes in levels of water and air pollution. The use of such instruments is possible with new platforms that combine multiple processor and wireless network modules. These platforms have energy control systems to allow the nodes themselves, and their sensor devices, to operate only on demand, thus conserving energy for long-life operations.
Sensors mounted on trams or other mobile platforms are allowing an innovative approach to the flexible and efficient deployment of environmental sensors. In what is termed actuated sensing, fixed sensors can communicate the local presence of an unusual dynamic condition (e.g. a frost or dew point condition or a rare bird call) to a mobile system, tasking it to move to scan the area to better understand the spatial and temporal scales of the phenomenon or animal presence.
Beyond sensors and sensor platforms themselves, however, other critical components of these new technologies consist of the means for coordination of sensor modalities across multiple spatial and temporal scales, the infrastructure to link sensors to a broadly accessible wireless network, and of course the reliability for long-term deployment with appropriate maintenance and calibration of sensors. Key to the success of these systems are appropriate tools for the storage and management of large data sets so that users can rapidly and efficiently access multiple configurations of data sets in real-time.
Remote sensing of ecological patterns and processes is a key element of the new technologies being applied to environmental monitoring and ecosystem model development. Resulting data are being collected by a combination of satellite sensors in earth orbit and instrumentation mounted in small aircraft. These sensors provide measurements of structural, spectral, and thermal characteristics of the land surface at a scale broader than that measured by fixed sensor arrays. Examples of such sensing instruments are multi-spectral imaging by the Moderate Resolution Imaging Spectroradiometer (MODIS), and radar interferometry, in satellites; and light detection and ranging LiDAR (laser altimetry), and thermal imagers, in aircraft.
MODIS is a key instrument aboard two NASA satellite systems. These satellites view the entire Earth’s surface every 1 to 2 days, acquiring data in 36 spectral bands, or groups of wavelengths. The multi-spectral sensing capability of MODIS allows it to quantify surface characteristics of the earth such as land cover type, snow cover, surface temperature, foliage cover of vegetation, and fire occurrence. These data also allow analysis of leaf area, leaf duration and net primary productivity at a landscape scale, and thus provide important inputs to parameterize or validate models of ecosystem sustainability. Data from satellite-based instruments such as MODIS are allowing scientists to improve our understanding of global dynamics and processes, and further to develop models to predict spatial and temporal scales of global change across the landscape. MODIS has also been used for other regional analyses in Southern California. For example, UCLA researchers have recently used satellite images from MODIS of the 2003 Southern California wildfires to assess the exposures of residents of the region to fine and coarse particles generated by the fires.
Synthetic aperture radar (SAR) interferometry is another satellite-carried instrument developed to detect subtle changes in the earth’s surface over periods of days to years with an unprecedented scale (global), accuracy (millimeter-level), and reliability (round-the-clock, all-weather). Over longer time scales of several years or more, high-resolution topographic data collected with SAR can also be used for large-scale change detection by comparing elevations at different times. This technique allows measurement of catastrophic changes in topography due to earthquakes, landslides, major floods, volcanic activity, and glacial melting. For example, SAR data has been effectively used to measure seismic displacement associated with earthquakes in California.
LiDAR instrumentation mounted in small aircraft provides the capacity for three-dimensional characterization of vegetation structure, and thus the structural complexity of forest stands (Figure 3). Unlike video sensors, lidar directly measures the distribution of vegetation structure along a vertical axis, and can provide measures of canopy height, stand basal area, biomass and total cover to a remarkable level of precision. Such data have wide application in forest and agricultural management. Key to this technology is the joint use of high-speed laser rangefinders, precise inertial navigation systems to measure the three-dimensional movement of the host aircraft, and paired GPS systems on the aircraft and a ground station for precise positioning.
As new technologies allow us to collect massive sets of data across broad geographical areas in a manner not previously possible, a critical challenge lies with how researchers and resource managers will manage and utilize such large masses of data. The goal, of course is to allow researchers to access these data streams in real time, to quickly analyze them, and to utilize models to apply complex data streams to help mitigate environmental problems. Many of the most significant questions related to the complexities of our environment lie at interfaces: the interface of atmosphere with soil systems, soil with freshwater aquatic systems, and freshwater with marine ecosystems. Understanding these complex interactions requires real-time linkages between data streams from sensor arrays operating in the air, in plant canopies, in the soil, and in adjacent waters.
Rapid progress in technologies for commercial wireless networking, now widely available to the general public, has provided an important advancement for networked sensors both in local area systems covering a few km2, as well as for regional systems extending over distances of 10-100 km. These WiFi technologies allow inexpensive, energy efficient and broadband connectivity through microservers to the Internet. Since the WiFi infrastructure is low cost and self-configuring, its deployment in natural and urban environments is convenient and rapid, accommodating wireless sensor arrays over a communication range of 100-200 m.
Commercial wireless technology also allows for long-range broadband links that may connect observation systems over large regions. One exciting example of new technologies for data access and transmission can be seen with HPWREN (High Performance Wireless Research and Education Network), a joint effort of the San Diego Supercomputer Center (SDSC) and the Scripps Institution of Oceanography. This exploratory project has created a high-performance, wide-area wireless network that spans much of San Diego County and adjacent counties. It includes backbone nodes on the UC San Diego and San Diego State University campuses, as well as a number of remote areas in San Diego County, including mountain peaks with hundreds of square miles of line-of-sight coverage. The HPWREN data communications infrastructure provides wireless high-speed Internet access for emergency data communications by local government agencies and first responders, for field researchers from many disciplines (geophysicists, seismologists, astronomers, oceanographers, and ecologists), and for rural Native American learning centers and schools. In using the high-speed HPWREN network, with a capacity to transmit 45 million data bits per second, emergency workers in mountain and desert locations and field researchers at remote sites can wirelessly transmit large amounts of data in real time. Because of the network’s high speed, images from high-resolution cameras can be instantly transferred over the network without interfering with other traffic.
The National Science Foundation (NSF) is in advanced planning stages for a major environmental program called the National Ecological Observatory Network (NEON). The mission of NEON (Figure 4) is to increase our understanding of how U.S. ecosystems and organisms respond to variations in climate and changes in land use at regional and continental scales. Understanding the significance of land use changes on our environment, and doing this in a manner relevant for urban planners and decision makers, provides a core component of the program. Land use changes that affect ecosystems and organisms include the conversion of land from wild to managed or urban land cover, and from agricultural uses to urban environments. The program will use new technologies, as described above, to measure important feedbacks between the biosphere and the atmosphere that are associated with alterations in land use, land cover, and vegetation. NEON will also investigate interrelationships between climate dynamics, biodiversity, invasive alien species, and emerging diseases such west Nile virus. Thus, NEON programs will greatly advance our regional efforts toward environmental sustainability for Southern California by allowing us to better understand the environmental implications of land use policies, and by helping to mitigate unwanted effects of global change.
As the influence of human activities continues to change the state of our environment and natural ecosystems, resource management efforts have responded by becoming far more interdisciplinary, integrative, and collaborative. Efforts to address the environmental "grand challenges," such as the effects of climate change and land use on our Southern California ecosystems, are promoting the coordination of standard measurement protocols and data management infrastructure across our region.
The key goal driving the development of new technologies for environmental monitoring continues to be an improved understanding of the complex behavior of ecological systems in a world with dynamic climate variation, and a means of predicting future environmental sustainability. These complexities of our ecological systems in Southern California arise not only from the dynamic nature of our physical and chemical environment, but also from our diverse biological systems and most especially our human societies. Effective predictive models for understanding ecological complexities and their reciprocal implications for human activities will depend on the collection of high-quality data, new approaches to synthesizing and visualizing these data across multiple spatial and temporal scales, and knowledge transfer to allow resource managers and land use planners to take advantage of these advances in a real-time collaborative manner.
Engineers, information technologists, statisticians, and ecologists are all working together effectively today in collaborative programs to advance the applications of new technologies for environmental monitoring and to improve our understanding of how variations in land use and climate influence ecosystem structure and function, as well as the consequences of these variations for society. The Center for Embedded Networked Sensing (CENS) at UCLA has been a focus of this research, with active participation of scientists and engineers from USC, UC Merced, UC Riverside, and Caltech. More information on this program is available at http://research.cens.ucla.edu.
Deborah Estrin is a Professor of Computer Science at UCLA, holds the Jon Postel Chair in Computer Networks, and is Director of the NSF-funded Center for Embedded Networked Sensing (CENS). Estrin received her Ph.D. (1985) in Computer Science from the Massachusetts Institute of Technology, her M.S. (1982) from M.I.T. and her B.S. (1980) from U.C. Berkeley. Professor Estrin chairs the Sensors and Sensor Networks subcommittee of the National Ecological Observatory Network (NEON) Network Design Committee, has been a co-PI on many NSF and Defense Advanced Research Projects Agency (DARPA) funded projects and has served on panels and committees for many networking related conferences.
William Kaiser is a Professor of Electrical Engineering at UCLA. In 1994, he initiated the first wireless networked sensor research with a vision of linking the Internet to the physical world through distributed monitoring. This research thrust continues with development of new robotic wireless sensor systems for environmental monitoring. Kaiser received his Ph.D. from Wayne State University in 1984.
Philip Rundel is Professor of Biology in the Department of Ecology and Evolutionary Biology at UCLA, and a senior investigator in the Center for Embedded Networked Sensing (CENS). He joined the faculty of UCLA in 1983 after 13 years on the faculty of UC Irvine. He is a plant ecophysiologist who has worked for many years on a variety of aspects of the ecology of mediterranean ecosystems of California and similar areas in the world. His work with CENS involves the development and application of new technologies in embedded networked sensing systems, mobile sensing platforms, and wireless communication systems that can be physically embedded in an environment and act to reveal events and processes that are invisible to more traditional observational technology. He teaches courses on California Ecosystems and Conservation Biology.