PhenoCam Images and Canopy Phenology at Harvard Forest since 2008
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
A collaborative research network (“PhenoCam”) to provide automated "near" remote sensing of canopy phenology across the northeastern US and adjacent Canada has been initiated. A pilot study (2006-2007) at the Bartlett Experimental Forest previously demonstrated the viability of tracking both spring green-up and autumn senescence based on relative changes in red, green, and blue (RGB) color channel brightness values extracted from networked digital camera (“webcam”) images. In 2008, we installed commercial-grade digital webcams at a dozen established research sites across the northeast, from Ontario and New York across to Maine, with most sites concentrated within a few degrees of 45 deg N. Two additional Midwestern sites were subsequently added to the network. At seven camera sites (Bartlett, Howland, Harvard Forest, Groundhog River, Chibougamau, University of Michigan Biological Station, and Morgan Monroe State Forest), ongoing measurements of carbon and water fluxes are being made with the eddy covariance method, which will enable us to directly link phenology to seasonal variation in ecosystem processes. The seasonal trajectory of a “greenness index” for Harvard Forest shows a rapid rise in greenness in spring (coinciding with ground observations of budburst), a gentle decline over the course of the summer, and then a rapid decline with autumn senescence. This network is providing a very rich (and unique) dataset on spatial and temporal patterns of canopy phenology in the across this region. Half hourly images are uploaded to a project web page (http://phenocam.sr.unh.edu), which also features additional information about the project, including a protocol for camera deployment, download tools (so that the imagery is available to a wider community), site locations and contact information, etc. Image processing routines are under development. Time series of greenness will be added to the archive when this is complete; at present, only images are being made available.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.002 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.011 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.028 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it