Analysis of crash characteristics on freeways with depressed medians in Southcentral Alaska.

Bham, Ghulam H.; Hafele, Brendan; Jones, Kaylin · 2017 · ROSA P / Alaska. Department of Transportation and Public Facilities

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study, conducted by researchers at the University of Alaska Anchorage and sponsored by the Alaska Department of Transportation and Public Facilities, addresses safety concerns regarding freeways with depressed medians in Southcentral Alaska. The primary research objectives were to evaluate crash characteristics during winter conditions, identify contributing factors for median intrusions (specifically cross-median and in-median crashes), and determine the role of depressed medians in rollover incidents. The motivation stemmed from the high severity of roadway departure crashes, which account for a significant portion of fatal crashes, and the unique challenges posed by Alaskan winter weather and median designs ranging from 24 to 64 feet in width. The methodology involved analyzing archived crash data from 2007 to 2012 for four major freeways: the Glenn Highway, New Seward Highway, Minnesota Drive, and Parks Highway. The dataset included 2,531 crashes, with 340 median crashes detailed through police reports. To supplement archived data, which often underreported property-damage-only incidents, researchers collected drive-through video data during winters 2013–2015 to capture median intrusions shortly after snow events. The analysis employed multinomial logistic regression (MLR) to identify crash contributing factors and Kernel Density Estimation (KDE) combined with a Crash Factor Measure to identify spatial hotspots. Key findings revealed that cross-median crashes were 2.5 times more likely to result in severe injuries compared to in-median crashes. Rollovers constituted 72.9% of all median crashes, with a higher fatality rate (2.4%) than non-rollover crashes (1.1%). Median slope analysis indicated that flatter slopes (6H:1V) were associated with a higher frequency of non-rollover crashes, suggesting they reduce rollover risk, while steeper slopes (4:1 and 5:1) made median traversal more difficult. Median width also influenced crash frequency; generally, crash frequency decreased as median width increased, though crashes were most frequent on medians between 26 and 40 feet. Regression models identified specific predictors: rollover crashes were associated with severe injuries, driver inexperience, horizontal curves, surface ice, and median widths between 26 and 40 feet. Cross-median crashes were linked to multiple-vehicle collisions, light trucks, post-sunset driving on lighted roadways, and pavement rutting. Spatial analysis identified nine top hotspots, primarily on the New Seward and Glenn Highways, and highlighted six interchanges requiring detailed examination. The study concludes that depressed medians serve as effective snow storage and impact absorption zones but present a trade-off between preventing cross-median collisions and inducing rollovers. The findings imply that median design must balance slope and width to mitigate specific crash types. Recommendations include improving the dissemination of real-time road condition information, enhancing driver training to address inexperience and overconfidence with four-wheel-drive vehicles, and conducting further research on median treatments suitable for Alaskan conditions. The study also suggests reviewing on- and off-ramp designs near identified hotspots and improving crash reporting accuracy to better inform future safety policies.

Key finding

Cross-median crashes were 2.5 times more likely to cause severe injury than in-median crashes, and rollover crashes accounted for 72.9% of all median crashes.

Methodology

dataset

Sample size: 340

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence
• crash typology
• demographic disparities
• naturalistic crash near crash

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes