Table 4
Summary of model selection criteria (AICc= Akaike's information criterion corrected for small sample sizes) for generalized additive models that were examined to explain variation in coregonine larval densities at our Lakes Geneva and Superior study sites with site-specific bathymetric depths (Depth in m), Julian Day, edible zooplankton density (EZ, ind.L−1), and wind vectors (u = east-west, v = north-south) loaded as potential explanatory variables. Only models with an ΔAICc less than 4 are summarized for Lake Superior and less than 1 for Lake Geneva. Variables that were included in each model have an associated approximate P-value whereas dashes indicate the variable was not included. Weight is scaled from 0 to 1 and estimates the probability that a given model is the best among all models considered (LogL= log likelihood).
| Lake | Depth | Julian day | EZ | U | V | Family | Df | logL | AICc | ΔAICc | Weight |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Geneva | <0.001 | 0.10 | – | – | <0.001 | NegBin(log) | 5 | –73.659 | 158.819 | 0.000 | 0.159 |
| <0.001 | 0.14 | – | <0.001 | – | NegBin (log) | 5 | –73.708 | 158.916 | 0.097 | 0.152 | |
| <0.001 | – | – | <0.001 | – | NegBin (log) | 4 | –74.740 | 158.975 | 0.157 | 0.147 | |
| <0.001 | – | – | – | <0.001 | NegBin (log) | 4 | –75.060 | 159.098 | 0.280 | 0.138 | |
| <0.001 | – | 0.07 | <0.01 | – | NegBin (log) | 7 | –69.763 | 159.271 | 0.452 | 0.127 | |
| <0.001 | 0.13 | 0.12 | – | <0.01 | NegBin (log) | 7 | –70.066 | 159.790 | 0.972 | 0.098 | |
| Superior | <0.001 | – | – | 0.01 | – | NegBin(log) | 9 | –50.608 | 123.999 | 0.000 | 0.500 |
| <0.01 | 0.87 | – | 0.02 | – | NegBin (log) | 10 | –50.371 | 125.754 | 1.755 | 0.208 | |
| <0.001 | – | 0.61 | 0.01 | – | NegBin (log) | 10 | –50.486 | 126.233 | 2.234 | 0.164 | |
| <0.01 | 0.64 | 0.43 | 0.02 | – | NegBin (log) | 11 | –49.980 | 127.567 | 3.567 | 0.084 |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.