Overview

Progress was made with new sensor deployments and data collection in three general areas of terrestrial plant ecology at the James Reserve: modeling the photosynthetic responses of the star moss, Tortula princeps; the photosynthetic responses to sunflecks of bracken fern, Pteridium aquilinum; and using imagers as plant phenology sensors with the pan-tilt-zoom tower cameras.

Approach

For the mosscam project, we are trying to increase the robustness of modeling photosynthesis using the color changes of the star moss and external sensors for temperature and light. For the bracken fern project, we are measuring the photosynthetic responses of the fronds to light flecks and then will be using a camera to record the light environment in order to estimate light levels the fronds are receiving. For the plant phenology project, we are using daily images of a meadow and target plants in order to eventually use automated processes to detect flowering and leafing events.

Systems/Experiments

The moss Tortula princeps undergoes changes in reflected visible light during cycles of drying and hydrating in the field and the MossCam Project has collected digital images of T. princeps at least daily since 2003. A publication in the International Journal of Plant Sciences by CENS investigators (Graham et al. 2006), related the color change with photosynthetic capacity and calculated carbon gain around a small rainfall event.

Bracken ferns at the James Reserve are an excellent model system for the utilization of CENS and NIMS technologies for science-driven applications. Data collection on bracken ferns at JR was begun in 2004 using imaging systems as well as environmental and physiological measurements to examine patterns of phenology and ecophysiology across light gradients at the forest edge.

There are four cameras mounted on towers at the James Reserve. During the summer of 2006, these cameras were scheduled to take images every day near sunrise and sunset. During late fall through winter, the schedule was changed to just take these images twice a week. The images taken include close-up images of target plants as well as scans of entire sections of meadows. More than 200,000 images have been collected since last summer and the flushes and blooms of native annuals was recorded.

Accomplishments

Mosscam: Following up on the success of using these images for predicting carbon uptake, additional sensors were added to the moss in order to correlate temperature and light values at the moss surface with air temperature and light measured nearby. Data were collected for 89 continuous days. These nearby data have been collected for several years and the next steps will be correlating these existing values with photosynthesis and the water content of the moss, as measured by the mosscam, for a refined estimation of photosynthesis of the moss. Light response curves measured from samples of moss taken from the James Reserve were made to examine the relationship between light and net photosynthesis.

Another set of laboratory measurements relating the ambient measurements at the James Reserve to photosynthetic rate of the star moss were conducted. To the left is a temperature response curve relating the percentage of maximum photosynthesis under ideal conditions to temperature.

The ambient measurements consisted of four thermocouples inserted into the moss just outside the view of the MossCam. Four photodiodes, calibrated against a Licor PAR sensor, were placed normal to the surface of the moss in similar locations. One day’s worth of measurements is shown to the left, as an example. Data are averages of the four sensors and of the air temperature sensor of the MossCam and a photodiode in a nearby nestbox.

Continued studies with Bracken ferns in 2006 involved further measurements of their responses to sunflecks, measurements of stomatal counts, and measurements of chlorophyll contents of shade and sun fronds. The continuing hypothesis is that the fern as a whole has competing goals of maximizing its carbon gain and minimizing its water loss through the acclimation of individual fronds to their local environments. Exposed fronds (sun-fronds) may be acclimated to conserve water as a priority over carbon gain and thus will have fewer stomata per chlorophyll content. Fronds occurring more in the understory (shade-fronds) may be acclimated for carbon gain at the expense of more water lost and thus will have more stomata relative to their more chlorophyll per unit area. The image to the left is of stomata on the underside of a fern frond with arrows pointing to the stomata. The numbers of stomata differed between sun and shad fronds, with sun fronds having 407 ± 14 stomata per millimeter (mean ± SD), and shade fronds having 57% of that for sun fronds. Chlorophyll content also varied, with sun fronds having 0.83 ± 0.07 mg per mg leaf tissue and shade fronds having 2.25 times that. Thus, the shade fronds have more stomata relative to their chlorophyll content and are thus less “concerned” with water loss than sun fronds. The response to sunflecks was also measured in the ferns using gas exchange techniques. The line figure to the left is of photosynthetic induction. Lines indicate a fully induced response to external CO₂ concentrations and the colored symbols represent internal CO₂ concentrations during photosynthetic induction. The intersection of the induction curve lower on the fully induced curve indicates more stomatal limitation on photosynthesis occurs. Sun fronds (high) have curves that intersect lower than the shade fronds (low), supporting evidence that induction in sun fronds is limited by stomatal opening and thus future multiscale predictions of photosynthesis will have to be tailored to different regions where fronds grow.

For the phenology project, in order to aid the human user in analyzing these extensive sets of images and tag them, the following visualization tool was created. The above figure is of a webpage that displays a selected image set over a specified time period. A collection of simple image processing tools are included to help filter out useful information.

The main image processing tool is a color filter applied to the images in an RGB, HSI, or HSL color space. Setting the min/max values of the individual color components removes any colors falling outside of the range, resulting in a black image with only the desired colors displayed. After this color filter has been optimized by the user, it can then be applied to the entire image set, resulting in a sub-sample of the image set. The webpage then shows the filtered images along with a “blob” count. This method is used as a rough flower counter (see image below).

Other tools in this interface are used to help improve contrast in the image and to make it easier to view images taken in bad lighting. For example, contrast stretching and brightness histogram equalization were implemented for this purpose, as seen on the screen capture below.

Future Directions

For the mosscam project, we need to create a calibration between the sensors that had been temporarily placed on the moss with those that are permanently installed and have been collecting data for the duration of the experiment. With this calibration, we will then be able to predict moss temperature and light levels for days in which image data exists. The result will be a much more sophisticated model of photosynthesis than for the previous publication and we expect to have the results published in a plant physiology or remote sensing journal.

For the bracken fern project, we will be employing the NIMS 1 camera to monitor light levels simultaneously with PAR measurements from fixed sensors. We will then make a seasonal comparison of instantaneous photosynthetic rates based on the camera data. Also planned is an experiment to limit the transport of photosynthates to the root system by cold girdling the fronds in order to get a better estimate of net primary productivity.

For the plant phenology project, we will be installing a serial connection to one of the cameras to gain more control of the shutter speed and aperture to improve the quality of the images of the meadow for 2007. Additionally, embedded computing will be employed to maximize the possibility of capturing blooming events and reducing unneeded image traffic through the James Reserve internet connection.

People

Staff:

• Eric Graham, UCLA
• Eric Yuen, UCLA
• Yeung Lam, UCLA