Representing uncertainty in a spatial invasion model that incorporates human-mediated dispersal
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
Most modes of human-mediated dispersal of invasive species are directional and vector-based. Classical spatial spread models usually depend on probabilistic dispersal kernels that emphasize distance over direction and have limited ability to depict rare but influential long-distance dispersal events. These aspects are problematic if such models are used to estimate invasion risk. Alternatively, a geographic network model may be better at estimating the typically low likelihoods associated with human-mediated dispersal events, but it should also provide a reasonable account of uncertainties that could affect perception of its risk estimates. We developed a network model that assesses the likelihood of dispersal of invasive forest pests in camper-transported firewood in North America. We built the model using data from the U.S. National Recreation Reservation Service, which document visitor travel between populated places and federal campgrounds across the U.S. and Canada. The study area is depicted as a set of coarse-resolution map units. Based on repeated simulations, the model estimates the probability that each unit is a possible origin and destination for firewood-facilitated forest pest invasions. We generated output maps that summarise, for each U.S. state and Canadian province, where (outside the state or province) a camper-transported forest pest likely originated. Treating these output maps as a set of baseline scenarios, we explored the sensitivity of these “origin risk” estimates to additive and multiplicative errors in the probabilities of pest transmission between locations, as well as random changes in the structure of the underlying travel network. We found the patterns of change in the origin risk estimates due to these alterations to be consistent across all states and provinces. This indicates that the network model behaves predictably in the presence of uncertainties, allowing future work to focus on closing knowledge gaps or more sophisticated treatments of the impact of uncertainty on model outputs.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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