Figure 11. (a) Map of Icy Point marine terraces showing area of lidar topography (Witter et al., 2017b) featured in (b) and the locations of two elevation profiles across Terraces A and B. (b) Lidar topographic map of the southern-most nose of Icy Point. Shore-parallel beach ridges visible on the lowest surface, Terrace A, evidence gradual, ongoing glacial isostatic rebound following the LIA. Erosional beach scarps delineate shorelines (B1, B2, etc.) beveled into Terrace B and elevated by uplift during repeated earthquakes. (c) Topographic profile F–F’ plotting the lowest lidar-measured elevations (black line) along a 50-m wide swath shown in map (a); all other elevations along the swath plotted in gray. Numbers below the profile indicate shoreline angle elevation in meters (NAVD 88). Beach sand below shoreline B8 dates to 3.7–7.0 ka. (d) Topographic profile G–G’ extending from the modern shoreline to the ancient seacliff separating Terrace B and Terrace C. Dune sand covering the highest shorelines dates to 3.7–5.8 ka.
4.6 Age of the paleo-seacliff separating Terraces B and C
We dated coastal landforms to estimate the age of the 30-to-60-m-tall paleo-sea-cliff that separates Terrace B from the older and topographically higher Terrace C (Figures 4 and 9). On the highest surface at the southern end of Terrace B, at an elevation of ~56 m near the foot of the paleo-sea cliff, we excavated a ~2-m-deep pit in an aeolian sand dune (Sand dune, Figure 4). The parabolic shape of the dunes, with steepest aspect toward the southwest, suggest katabatic winds blowing off glaciers to the northeast mobilized beach sand and deposited it on emergent marine terraces. At depths of 1.05 and 1.42 m in the pit we sampled aeolian sand for IRSL age determinations on feldspar grains to estimate the last time the dune sand had been exposed to light. The shallower sample has an IRSL age of 4.18 ± 1.39 ka and the deeper sample has an IRSL age of 4.74 ± 1.10 ka (Table 1).
We also sampled beach sand from an exposure of marine terrace sediment below Terrace B shoreline B8 at an elevation of ~35 m near the top of a steep-sided ravine (Tocher’s Gulch, Figures 4 and 11). The beach sand deposit was ~4 m thick above Topsy Formation bedrock and we sampled at depths of 2.70 m and 3.89 m below the terrace surface. The shallower sample has an IRSL age of 4.39 ± 0.66 ka and the deeper sample has an IRSL age of 5.99 ± 1.02 ka (Table 1). From these ages and their uncertainties we infer that beach processes began depositing sand before 5.0–7.0 ka and terrace emergence occurred sometime after 3.7–5.0 ka.
The IRSL ages of separate landforms provide independent minimum estimates for the time of abandonment of the highest shoreline angle that separates Terrace B from Terrace C (Figures 4 and 12). The highest shoreline angle was abandoned before 5.0-7.0 ka because beach sand exposed in Tocher’s Gulch, at an elevation more than 20 m below the highest shoreline angle, was active at this time. A second minimum age for the emergence of the paleo-sea cliff comes from the age of aeolian sand, 3.6–5.8 ka, which advanced over Terrace B after its highest shoreline was abandoned. At a human occupation site near Tocher’s Gulch, charcoal at the base of aeolian sand dated to 4.8–5.2 ka (Crowell et al., 2013). Considering the IRSL data and the coastal landforms from which the samples come, the minimum age of 5.0–7.0 indicated by beach sand more closely limits the time of abandonment (Figure 8).
To estimate the age of the Terrace B shoreline angle we use a numerical model (OxCal, v4.4.2; Bronk Ramsey, 2009; 2023) to compute the 95% age range (probability density function) based on maximum and minimum age constraints. Radiocarbon ages from outwash gravel graded to the highest Terrace B shoreline provide maximum age constraints that range from 9.1 to 10.1 ka. Minimum age constraints come from younger beach deposits, estimated to range from 5.0 to 7.0 ka, that post-date the cutting of the Terrace B shoreline angle. Given these age constraints, the shoreline angle at the top of Terrace B was cut between 5.5 and 9.4 ka (Figure 8).
4.7 Reconstructing the tectonic component of relative sea-level change at Icy Point
Profoundly different Holocene RSL histories occur on either side of the Fairweather fault. Although geomorphic features left by the LIA sea-level highstand do not vary across the fault, tectonic features differ markedly across the fault. Based on these differences evident within hundreds of meters of the Fairweather fault, we show that the different RSL histories can be explained by tectonic uplift west of the Fairweather fault (Figure 12).
In Icy Strait, 30 km east of the Fairweather fault, RSL change fluctuated from 4.1 ±1 m below present sea level at 6.9–7.2 ka to its present elevation, including the Little Ice Age high stand that registered ~4 m above present for the several centuries ending by 1750–1800 CE (Mann and Streveler, 2008) (Figure 12). Although Mann and Streveler (2008) assume that tectonic effects were unimportant in the RSL history in Icy Strait east of the Fairweather fault, in fact their reconstruction incorporates all of the varied factors affecting RSL: eustacy, isostacy, tectonism, local, and other processes (Shennan et al., 2012). We adopt their assumptions and utilize their Icy Strait RSL reconstruction 30 km northeast along the open-ocean coast from Icy Strait to the east side of the Fairweather fault at Palma Bay (Figure 1). Several observations support this extension: 1) relatively low thermochron ages for rock samples at Palma Bay sites east of the Fairweather fault indicate low rock uplift (Lease et al., 2021): 2) the absence of marine terraces along the coast of Palma Bay; 3) low seismicity and the lack of mapped convergent structures east of the Fairweather fault at the latitude of Icy Point; and 4) geodetic data that indicate that fault-normal strain can be accommodated by reverse faults west of the Fairweather fault.
In contrast to the coastline east of the Fairweather fault, RSL at Icy Point has fallen 42.3 ±2.1 m during the Holocene based on the elevation of the Terrace B shoreline angle relative to the modern shoreline (Figure 12; Table 4). We reconstruct RSL change at Icy Point by comparing the heights of the Terrace B shoreline angle with its modern equivalent. The height of the Terrace B shoreline angle was estimated by subtracting the thickness of terrace sediment (3 ±1 m), approximated from field exposures and pits, from the elevation of the inner edge of terrace B at the base of the sea cliff leading up to Terrace C. We determined the mean elevation (47.3 ±1.7 m) of the Terrace B inner edge by tabulating the elevation of the sharpest inflection at the base of the sea cliff in 25 lidar profiles (Table 4). The elevation of the modern shoreline angle (2.0 ±0.8 m, NAVD 88 datum) was determined from measurements of 50 lidar profiles along the inner edge of the shore platform. We subtract the Terrace B shoreline angle from the modern shoreline angle (Table 4) to estimate the change in RSL at Icy Point (-42.3 ±2.1 m) since the original formation of Terrace B between 5.5–9.4 ka.
To evaluate the contribution of tectonic processes to RSL changes at Icy Point, we assess the difference between Holocene RSL change on either side of the Fairweather fault (Figure 12), including all factors that affect RSL change (Shennan et al., 2012; Shennan, 2015). We assume that, within errors of the methods we use and those of Mann and Streveler (2008), the contributions of eustacy, isostacy, and local tidal effects to RSL change do not vary directly east and west of the Fairweather fault at Icy Point. However, the tectonic contribution to RSL change varies substantially across the fault. Therefore, taking the difference in Holocene RSL change across the fault results in a measure of the tectonic contribution of RSL.
To assess the difference in RSL change across the Fairweather fault, we first reconstruct Holocene RSL curves for Icy Point and Palma Bay (Figure 12). We estimate that since the formation of the Terrace B shoreline angle at Icy Point between 5.5–9.4 ka, RSL has fallen dramatically, equaling -42.3 ±2.1 m (Table 4); whereas, based on reconstructions along Icy Strait, including Palma Bay directly east of the Fairweather fault, RSL has risen +4.1 +1/-0.75 m since 6.9–7.2 ka (Mann and Streveler, 2008). The difference in RSL change across the Fairweather fault indicates that the tectonic contribution to RSL change at Icy Point is -46.4 ±2.4 m (Table 4, Figure 12). The rate of RSL fall at Icy Point, -6.8 ± 2.2 m/ky, is the quotient of the tectonic contribution to RSL change divided by the age of the Terrace B shoreline angle (see Figure 4). We attribute this substantial rate of RSL fall at Icy Point to tectonic rock uplift since the formation of Terrace B at the rate of 6.8 ± 2.2 m/ky (4.6 to 9.0 mm/yr).