Strengths

• Long-term continuous record of annual snow melt onset dates on Arctic sea ice since 1979. The 43-year record is useful for monitoring climate trends and inter-annual variability (Bliss & Anderson, 2014; Box et al., 2019; Markus et al., 2009).
• Weather filtering implemented in the AHRA algorithm reduces effects of short-term weather interference in melt onset detection (Drobot & Anderson, 2001).
• The melt onset dates are fully reprocessed with the most up to date source data. Source data from different sensors are fully inter-calibrated for a consistent time series (Bliss, forthcoming).
• Newly implemented algorithm start date methodology improves melt onset date detection on the ice periphery, better reflecting variability in the annual sea ice growth cycle (Bliss et al., 2017; Bliss, forthcoming).

Limitations

• Only one melt onset date is archived per year at each sea ice grid cell, thus, any freeze/thaw conditions that occur following the melt onset date are not captured.
• Passive microwave-based melt onset date data products have higher uncertainties in regions of thin ice or low sea ice concentration where open water may be mistaken for melt (Bliss et al., 2017).
• Low spatial resolution of the gridded data (nominally 25 km) make it difficult to ground truth melt onset dates with in situ observations or cross-compare with other data sources (Bliss & Anderson, 2018).