Comparing generalized and customized spread models for nonnative forest pests
Bibliographic record
Abstract
While generality is often desirable in ecology, customized models for individual species are thought to be more predictive by accounting for context specificity. However, fully customized models require more information for focal species. We focus on pest spread and ask: How much does predictive power differ between generalized and customized models? Further, we examine whether an intermediate "semi-generalized" model, combining elements of a general model with species-specific modifications, could yield predictive advantages. We compared predictive power of a generalized model applied to all forest pest species (the generalized dispersal kernel or GDK) to customized spread models for three invasive forest pests (beech bark disease [Cryptococcus fagisuga], gypsy moth [Lymantria dispar], and hemlock woolly adelgid [Adelges tsugae]), for which time-series data exist. We generated semi-generalized dispersal kernel models (SDK) through GDK correction factors based on additional species-specific information. We found that customized models were more predictive than the GDK by an average of 17% for the three species examined, although the GDK still had strong predictive ability (57% spatial variation explained). However, by combining the GDK with simple corrections into the SDK model, we attained a mean of 91% of the spatial variation explained, compared to 74% for the customized models. This is, to our knowledge, the first comparison of general and species-specific ecological spread models' predictive abilities. Our strong predictive results suggest that general models can be effectively synthesized with context-specific information for single species to respond quickly to invasions. We provided SDK forecasts to 2030 for all 63 United States pests in our data set.
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.
How this classification was reachedexpand
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".