Forecasting the Inland Penetration of Lake-Effect Long-Lake-Axis Parallel Snowbands

Successful forecasting of lake-effect long-lake-axis parallel (LLAP) snowbands is crucial for public safety, since LLAP bands are well-known to produce hazardous weather downwind of the parent lake.  While previous studies have focused primarily on LLAP bands over-lake or relatively close to shore, LLAP bands are also known to extend hundreds of km inland on occasion, making the inland penetration of LLAP bands an important safety concern.  However, the current conceptual model for LLAP bands does not fully explain their inland penetration.   While this model accounts for the large production of energy by cold air traversing a relatively warm lake surface (long-considered to be the main energy source for lake-effect circulations), it does not suggest any additional power source for the inland segments of LLAP bands.  Thus, the conventional model makes a tacit assumption that LLAP circulations overturn once following landfall, thereby expending their residual buoyancy, and then disintegrate.

In contrast to this prediction, an examination of satellite imagery, undertaken in preparation for the Ontario Winter Lake-effect Systems experiment (OWLeS), revealed multiple cases with great persistence of inland convection.  Analysis of LLAP cases observed during OWLeS confirmed this observation.  These findings indicate the existence of a secondary power source for LLAP bands.  The nature of this power source was suggested by the observation that inland penetration of LLAP bands is often greatest during the period of cold advection immediately following cold frontal passage.  Later in the synoptic cycle conventional measures of lake-effect forcing, such as air/lake temperature difference, often attain higher values, but inland penetration is nonetheless reduced.

While the presence of cold advection suggests positive surface heat fluxes (as the surface equilibrates with the overlying boundary layer), further analysis revealed that overland surface heat fluxes are incapable of statistically explaining the inland penetration of LLAP bands.  This finding indicates a significant difference between the forcing mechanisms that generate LLAP bands and those that maintain them once over land.  Examination of the North American Reanalysis data and NEXRAD reflectivity maps revealed that the extent of inland penetration of significant snowfall is correlated with differential thermal advection.  Cold advection increasing with height acts as a local power source since it destabilizes the corresponding atmospheric layer.  We hypothesize that this power production extends the inland longevity of existing LLAP bands.  The power supplied by differential thermal advection (DTA) is related to both the depth of the DTA layer and the magnitude of DTA.  Accordingly, regression was used to develop a new metric for predicting the inland penetration of Lake Ontario LLAP bands based on boundary layer depth and thermal advection near the boundary layer top (Z_i).  This new metric requires only inputs readily available to forecasters from synoptic-scale models.  The efficacy of this new metric was tested alongside indicators traditionally used by forecasters to predict LLAP bands (such as the potential for lake-effect instability and boundary layer windspeed) on 37 instances of LLAP bands.  The results of this test showed that traditional LLAP predictors were unable to accurately predict the inland penetration of LLAP bands while the combination of boundary layer depth and near-Z_i thermal advection produced significant skill.