OVERVIEW

Sensors and sensor networks provide coordinated, simultaneous, multi-modal observations of small-scale temporal and spatial ecosystem processes. For example, to project CO2 flux from a single satellite image pixel to the North American continent, one will scale 11 orders of magnitude. But, respiration and growth rates are determined by nutrient exchange and respiration processes that occur within an area of only a few _m, on the surfaces of bacterial colonies, fungal hyphae, and at the interface of the bulk soil and the rhizosphere. Thus, scaling from the actual process to the pixel is actually 14 orders of magnitude. In our preliminary tests, we have shown that there is as much variation in processes such as CO2 production and sequestration within a few m2 of land surface as exists across biomes. Just as important, shifts in process rates change dramatically as a wetting front moves in a fractal pattern down through the soil following precipitation. Respiration increases actually precedes the wetting front. Yet, how do we integrate the information into a framework that can make real-time assessments of processes? This approach transforms our understanding of the mechanisms regulating larger-scale ecosystem change. We are currently emplacing the initial array of sensors coupled to minirhizotron cameras that will allow us, for the first time, to simultaneously and in situ observe roots, fungal hyphae, and bacterial colonies at the same time and place as measurements of CO2, temperature, moisture (and soon, NO3- production) are undertaken. Thus, the widespread use of Embedded Networked Sensing (ENS) to achieve integrated, cross-scale (multi-scale) observations will ultimately enable mechanistic-based scaling and forecasting of ecosystem processes from the microbial to the biome level. In this way, ecosystem state will be observable as a dynamic response to gradients and environmental transients rather than as an equilibrium described by average properties. Real-time data access allow the researcher to review and assimilate data, explore online models while in the field, and to guide adaptive measurements, sampling, and calibration of sensors. Research contributions from the Terrestrial group at CENS now explore how such access can transform ecological experimentation from a batch process into an interactive one, thereby increasing the rate and scope of ecological exploration. The core activity is to develop essential infrastructure in support of ENS for ecological observatories, such as CLEANER and NEON, and in related technology transfer applications such as precision agriculture and centers of active ecological research. The Center's field test bed located at the University of California, James San Jacinto Mountains Reserve, in collaboration with other CENS groups, has successfully deployed several unique, overlapping, multi-scale embedded sensor arrays, at fixed locations and on mobile platforms.

Most Significant Accomplishments

Terrestrial Ecology Observing Systems (TEOS) applications group: During the third year our group has broadened its focus from habitat-related to ecological observing systems by expanding our test bed to incorporate a continuously operating, heterogeneous array of mobile and fixed microclimate sensors, coordinated mutli-scale imagers, and other specialized distributed instrumentation, a database archive and web interface for queries, scientific data visualization, and data export functionality. These systems are being used for multi-scale studies from root and fungal respiration, plant ecophysiology and growth responses to microclimate, tree canopy monitoring, and avian research. Extensive planning and engineering design has been accomplished towards future data portals and coordinated modeling approaches, technology transfer and collaborations to wider the range of applications within temperate and tropical environments.

Research Objectives

The following are summaries of the objectives/accomplishments of our primary projects.

1. Microclimate, Acoustic, and Image Acquisition Sensor Networks:
   • Micro Climate Sensor Array: Doubled the number of microclimate sensor sites and continued to operate the original wireless sensor sites, starting on their third year of data collection.
   • Automated Image Acquisition: Expand the image collection system to 13 nest boxes and 4 network controlled tower-mounted cameras. Time-lapse image sequences are collected automatically, documenting nesting activity and nest success, plant phenology and ground conditions (snow cover).
   • Measurement and Image User Interface: Create a web interface for interactive browsing of measurements (graphs and raw data) as well as browsing time sequences of images.
   • Initial Biological Investigations: The web interface for James Reserve data access has made possible an initial investigation correlating microclimate data with spatially resolved spectroscopy of image data and has resulted in a project modeling the annual carbon budget of local bryophyte communities. The imaging of phenology and photosynthetic status of plants with different growth forms (Bracken fern, hardwood Rhododendron, Star moss) has been successfully demonstrated and is now part of continuing and new experimental investigations. The appropriate embedded networked sensors and microclimatic data required for measuring and correlating with water movement in various hardwood species was investigated and is now incorporated into the design of current ecological experiments. Progress continues on the expansion of our nest box monitoring systems to study avian reproductive success.
   • Adaptive Communication in Sensor Arrays: design of an acoustic sensor array that learns to separate the nuances in vocalizations of bird species, initially within Acorn Woodpeckers, but ultimately a wide range of vertebrates. The system would differentiate individuals as well as localize in 3-space the calls of animals using a combination of acoustic processing and video surveillance. Fieldwork proceeds at both the James Reserve and the Hastings Natural History Reservation.
2. Embedded network sensing within soils
   • Soil Array Laboratory Test Bed: Adaptation of proven soil moisture sensing techniques has allowed the development of temperature-insensitive micro-sensors to be used in the soil monitoring test bed. This laboratory test bed is being used to investigate the effect that diurnal temperature changes have on water movement through soil and how that water movement is affected by the presence of rocks on the soil surface.
   • AMARSS: Development of the Automated Minirhizotron Array with Rhizosphere Soil Sensors (AMARSS) has proceeded with installation of minirhizotron tubes and testing of soil water potential, moisture, temperature and CO2 sensors. Engineering design and development is underway for robotically controlled imaging systems.
   • Tropical Agricultural test bed: An ethnobotanical study underway in French Polynesia is using micro-climate and soil sensor arrays to evaluate the influence of weather conditions and the synodical lunar rhythm on the success of local sweet potato crops (Ipomea batatas) as predicted by traditional Tahitian agricultural and ecological knowledge.
3. Research Infrastructure:
   • Utilities and Networks: Maintained the James Reserve's local and wide-area network system (wired and wireless), data storage systems, and our locally generated electrical power distribution systems. Extended power, video and Ethernet connectivity to 14 sites around the Reserve.
   • Major Construction: Construct three new 10-meter fiberglass towers at key observation points around the Reserve and installed the suspension cable and related systems for the NIMS2 transect.
   • Maintenance and Operations: A severe lightning storm this year damaged many components of the distributed video monitoring system and some key components of the power generation and distribution system. All of these have been repaired or replaced.
   • Baseline Remote Sensing Data: Complete 10cm resolution aerial photography of the James Reserve and Hall Canyon Research Natural Area.
   • Experimental System Support: Support 3 deployments of experimental wireless sensor systems to evaluate RF communication hardware, communication protocols, sensor interface hardware, and user interfaces (includes ESS2).

Future Directions and Planning Activities

Continued refinement of applications of CMS, ESS, and NIMS technologies will be across larger spatial scales, incorporating new and different sensors and sensing protocols. The synthesis of data from multiple sources and scales is essential for ecological studies and support for multi-modal investigations by providing open and easy access to the different types of image and data streams is top priority. Current investigations, such as modeling carbon budgets based on spatially-resolved spectroscopy of image data and correlating the imaging of phenology with sap flow measurements and micrometeorological conditions, require such multiple-source data with diverse sets of data access needs. Improvements in CMS, ESS, and NIMS technologies will facilitate such ecological investigations and create, as an emergent property, the possibility for entirely new modes of ecological investigations. CENS Terrestrial researchers hosted an important workshop on sensor array networks in tropical ecosystems. This workshop resulted in a major instrumentation proposal submitted to NSF to develop an ecological observatory for the La Selva Biological Station in Costa Rica. Significant progress was made in the development of the EMISSARY sensor array interface, which will integrate soils data collected into a geospatial modeling framework, using a micro-GIS system for managing and displaying and analyzing the data. The baseline microsensornets development for system-wide ecological research will be to the other UCR Natural Reserve sites and a NIMS sensor array will be implemented at the Wind River Canopy Crane Research facility and the La Selva Biological Station, Costa Rica. In addition, CENS personnel continued to participate in the NSF NEON National Design Consortium and Regional EON activities.

Problems Encountered

No unanticipated significant problems have been encountered. Equipment failures, and protracted software development are normally expected for this type of engineering applications. This information helps to improve design and ultimately expand functionality of our various systems.