Plant phenology relates strongly to primary productivity and the energy that enters into ecological food webs, and thus is vital in understanding ecosystem function and the effects of climate and climate change. The manual collection of phenological data is labor-intensive and not easily scalable, thus the ability to quantify leaf flush and other parameters at many locations requires innovative new methodologies such as the use of visible light digital cameras. Improved imaging performancewas obtained by using a cabled, mobile camera system that allowed a repeated image census of branches of Rhododendron occidentale in the understory along a 30m transect during leaf flush. Automatic division of acquired images into areas of interest (leaves) and background for calculating leaf area was accomplished by thresholding images in different color spaces. Transformation of the color space into the hue, saturation, and luminance (HSL) color space before thresholding resulted in a mean RMS error of 21.2 cm2 compared to hand-counts of leaf area. Thresholding in the native red, green, and blue (RGB) color space to isolate leaves resulted in a larger error, as did using algebraic combinations of the color components or color ratios. Relating physiological function to images, as for sap flow for branches of R. occidentale, indicates that the greening and calculated leaf area of a species as detected by imagers requires additional meteorological sensor data for interpretation.

Environmental sensor networks offer a powerful combination of distributed sensing capacity, real-time data visualization and analysis, and integration with adjacent networks and remote sensing data streams. These advances have become a reality as a combined result of the continuing miniaturization of electronics, the availability of large data storage and computational capacity, and the pervasive connectivity of the Internet. Environmental sensor networks have been established and large new networks are planned for monitoring multiple habitats at many different scales. Projects range in spatial scale from continental systems designed to measure global change and environmental stability to those involved with the monitoring of only a few meters of forest edge in fragmented landscapes. Temporal measurements have ranged from the evaluation of sunfleck dynamics at scales of seconds, to daily CO2 fluxes, to decadal shifts in temperatures. Above-ground sensor systems are partnered with subsurface soil measurement networks for physical and biological activity, together with aquatic and riparian sensor networks to measure groundwater fluxes and nutrient dynamics. More recently, complex sensors, such as networked digital cameras and microphones, as well as newly emerging sensors, are being integrated into sensor networks for hierarchical methods of sensing that promise a further understanding of our ecological systems by revealing previously unobservable phenomena.

The characterization of the solar radiation environment under a forest canopy is important for both understanding temperature-dependent biological processes and validating energy balance models. A modified sinusoidal model of soil heat conductivity was used to estimate subsurface temperature and heat flux from the uneven but periodic solar heating of the soil surface due to sun flecks from a forest canopy. Using a mobile sensor platform with an infrared thermometer along an 11 m transect, a sunfleck model of soil surface temperature was tested using soil surface temperature maxima, air temperatures, and photodiodes placed on the soil surface to measure sunflecks. A pan-tilt-zoom digital camera on a 10 m tower above the site was then used to capture a time series of panoramic images of sunflecks reflected from the soil surface and to scale the sunfleck temperature model to a wide area. Finally, this image-based model of surface temperatures was combined with the modified sinusoidal model for heat conduction to estimate soil subsurface temperatures and heat flux over a wide area due to sunflecks from a forest canopy.

There exist many natural phenomena where direct measurement is either impossible or extremely invasive. To obtain approximate measurements of these phenomena we can build prediction models based on other sensing modalities such as features extracted from data collected by an imager. These models are derived from controlled experiments performed under laboratory conditions, and can then be applied to the associated event in nature. In this paper we explore various different methods for generating such models and discuss their accuracy, robustness, and computational complexity. Given sufficiently computationally simple models, we can eventually push their computation down towards the sensor nodes themselves to reduce the amount of data required to both flow through the network and be stored in a database. The addition of these models turn in-situ imagers into powerful biological sensors, and image databases into useful records of biological activity.

Stationary and mobile ground-based cameras can be used to scale ecological observations, relating pixel information in images to in situ measurements. Currently there are four CENS projects that involve using cameras for scaling ecological observations: 1. Scaling from one individual to the landscape. Pan-Tilt-Zoom cameras can be zoomed in on a tight focus on individual plants and parts of individuals and then zoomed out to get a landscape view, composed of the same and similar species. 2. Estimating photosynthesis over large areas with HDR. High Dynamic Range imaging is a technique to capture an absolute amount of reflected light in an image. For a meadow composed of similarly reflecting species, we can estimate light received by leaves and thus photosynthesis over a wide area. 3. Scaling soil surface temperature measurements. Soil surface temperatures and soil energy balance are related to solar radiation and air temperature. Sunflecks captured with a camera taking panoramic mosaics of images can be used to estimate the radiation load for large areas of unobstructed understory. 4. Expanding plant phenological observations with a nation-wide network of webcams. Twice-daily images from over 1000 internet-connected and freely available cameras have been collected since February 2008. The advance of Spring can be tracked as a "green-up" and related to satellite remote sensing signals.

There exist many natural phenomena where direct measurement is either impossible or extremely invasive. To obtain approximate measurements of these phenomena we can build prediction models based on other sensing modalities such as features extracted from data collected by an imager. These models are derived from controlled experiments performed under laboratory conditions, and can then be applied to the associated event in nature. In this paper we explore various different methods for generating such models and discuss their accuracy, robustness, and computational complexity. Given sufficiently computationally simple models, we can eventually push their computation down towards the sensor nodes themselves to reduce the amount of data required to both flow through the network and be stored in a database. The addition of these models turn in-situ imagers into powerful biological sensors, and image databases into useful records of biological activity.

Ecological observatories are a new class of multiuser research infrastructure designed and deployed to address a broad range of continental-scale ecological questions that until only recently were not technologically feasible. These highly networked ecological observatories, spread across the United States and featuring a diverse integration of programmable sensing capabilities and remote observational functions are expected to enable a transformation in the scope of environmental research, particularly in relation to understanding how global climate shifts and local and regional land use changes will quantitatively affect the composition, structure, and dynamics of the nation’s ecosystems and services. Observing systems research focused on terrestrial ecology is one of four core research application fields of the Center for Embedded Networked Sensing, which operates an engineering and experimental test bed located at the James San Jacinto Mountains Reserve, a biological field station that is part of the University of California Natural Reserve System. This paper describes the various experimental and deployed embedded networked sensing systems at the James Reserve, and suggests how similar systems and related infrastructure will be key to meeting the engineering and science objectives of future ecological research.

PASEO was an NSF-funded PASI, which is a two-week training program for U.S. and Latin American early-career scientists and engineers. It took place in February 2009 in Bahia Blanca and Buenos Aires, Argentina. The host institutions were Instituto Argentino de Oceanografia (CONICET-funded), and Instituto Nacional de Tecnologia Industrial (INTI). Collaborating U.S. groups included GLEON, WATERS Network, and Codar Inc. The topic of PASEO was developing and deploying current sensor technologies to obtain a fuller understanding of environmental and ecological systems. PASEO was intended to be a multi-cultural and multi-disciplinary active learning experience. Applicants participated in active learning sessions related to the fabrication, deployment, and analysis of data streaming from environmental sensors. Scientific topics included lake metabolism, eco-hydrology in a saltwater marsh, soil moisture and energy balances, and plant phenology. Opportunities for hands-on training included buoyed and robotic sensor deployment, soil sensors, DTS temperature measurements, coastal radar, and low-temperature co-fired ceramics (LTCC) sensor fabrication methods.