Impact of Model Resolution on the Representation of the Wind Field Along Nares Strait

Abstract

Nares Strait is a major pathway along which multi-year sea ice leaves the Arctic, an ice class that seen a recent dramatic reduction in extent. The winds that blow along the strait play an important role in modulating this ice export as well as in establishing the Arctic’s largest and most productive polynya, the North Water, that forms at its southern terminus. However, its remote location has limited knowledge of the winds along the strait. Here we use automatic weather station from Hans Island, in the middle of the strait, to assess the ability of a set of atmospheric analyses with a common lineage but with varying horizontal resolution to represent the variability in the wind field. We find that the flow is highly bidirectional, consistent with topographic channeling, with the highest wind speeds from the north and that a model resolution of ~9km is required to capture the observed variability. The wind field at Hans Island is also found to be representative of variability in the flow along much of Nares Strait.

Impact of Resolution on the Representation of the Mean and Extreme Winds Along Nares Strait

“…The results presented here are consistent with previous studies (Kohnemann & Heinemann, 2021; Moore, 2021; Moore & Våge, 2018; Samelson & Barbour, 2008) that indicate that the representation of the flow along Nares Strait requires models with sufficient horizontal resolution to represent the ageostrophic processes that are the result of the narrowness of the strait and the steep topography that abuts it on both the Ellesmere Island and Greenland sides. Consistent with the results of Moore (2021), the ERA5 is unable to represent the topography of the region (Figure 2) and underestimates both the mean and extreme wind conditions at the stations in the vicinity of the strait (Table 1 and Figures 3 and 4) as well as the wind field along the strait (Figures 5 and 6). There are also improvements in the representation of the wind climate in the vicinity of Nares Strait in the CARRA as compared to the ECOA.…”

“…Winds, with either direction, in excess of 30 ms −1 were observed. In agreement with earlier work (Samelson et al., 2006), northerly flow at Hans Island was associated with higher sea‐level pressures over the Lincoln Sea as compared to northern Baffin Bay with the opposite being the case for southerly flow (Moore, 2021). Moore (2021) also noted that the winds at Hans Island were representative of the variability in the winds along much of Nares Strait.…”

“…The exception is Hans Island which is a small isolated island with a height of approximately 170 m above sea‐level. None of the models resolve the island and hence the 10 m winds are biased low (Moore, 2021). For this reason, we will also compare the 100 m winds from the models to the observations at the Hans Island site.…”

Estimation of Sea Ice Production in the North Water Polynya Based on Ice Arch Duration in Winter During 2006–2019

“…A similar low heat flux (∼189 W/m 2 in Figure 6b) was estimated in December 2012, because of the low wind speed (∼4–8 m/s) and high air temperature (∼250–255 K). However, it is worth noting that although the ERA5 model data is able to capture the large‐scale structure of these wind events including the pressure gradient at the Nares Strait, it can underestimate the high winds at the Smith Sound that are observed at the atmospheric weather station (AWS) due to its coarse spatial resolution, especially the extreme wind events (Gutjahr & Heinemann, 2018; Kohnemann & Heinemann, 2021; Moore, 2021; Moore & Våge, 2018).…”

“…The ERA5 has the higher resolution in the currently available long‐term reanalysis data, and the ERA5 model has an improved representation of the Nares Strait's topography but it still lacks some of the fine‐scale structure, including details along the Kennedy Channel and Smith Sound, thus the ERA5 data can underestimate the winds along the Nares Straits (Moore, 2021; Moore & Våge, 2018). The ECOA data can capture the wind speed better than the ERA5 data because of its higher resolution (∼9 km; Moore, 2021; Moore & McNeil, 2018; Moore & Våge, 2018), however, it does not have publicly available long‐term series data at this time. Therefore, the ERA5 data is currently the best data available to estimate heat flux and IP in the NOW over the period 2005/2006–2018/2019.…”