Satellite observations of asymmetrical physical and biological responses to Hurricane Earl

Authors

  • Wei Shi,

    1. Center for Satellite Applications and Research, National Environmental Satellite, Data, and Information Service, NOAA, Camp Springs, Maryland, USA
    2. CIRA, Colorado State University, Fort Collins, Colorado, USA
    Search for more papers by this author
  • Menghua Wang

    1. Center for Satellite Applications and Research, National Environmental Satellite, Data, and Information Service, NOAA, Camp Springs, Maryland, USA
    Search for more papers by this author

Abstract

[1] Asymmetrical physical and biological responses to Hurricane Earl in 2010 are revealed with a combined data set of the Moderate Resolution Imaging Spectroradiometer (MODIS) ocean color and Advanced Microwave Scanning Radiometer EOS (AMSR-E) SST observations onboard the satellite Aqua. Hurricane Earl induced broad SST drops and elevated chlorophyll-a concentrations along its track. The ocean's physical and biological responses are notably right-biased when the hurricane passed along the U.S. East Coast. In the ranges within 100 km off the track, the SST dropped 1.85°C and 1.23°C on the right and left sides, respectively. On the other hand, the ratios of the chlorophyll-a concentration before and after the passage of Hurricane Earl are 2.04 on the right side and 1.33 on the left. In addition to the satellite-observed sea surface changes, temperature and salinity profiles of an Argo float on Earl's track show the ocean's physical response occurred mostly within the mixed layer and thermocline in the upper 70 m water column.

1. Introduction

[2] It has been long known that the ocean experiences significant physical responses following a moving hurricane. Strong vertical entrainment and upwelling caused by hurricane winds can result in a couple of degree sea surface temperature (SST) drop and remarkable thermocline responses along the hurricane track [Brink, 1989]. Satellite SST measurements have documented synoptic SST drops following the passage of a hurricane [Monaldo et al., 1997]. Normally the SST can drop ∼2–4°C depending on the size, intensity and transit time of a hurricane. The negative anomaly SST will disappear in a time frame of ∼5 to 20 days [Price et al., 2008].

[3] The enhanced vertical mixing and upwelling also bring nutrient-rich deep ocean water or bottom sediments to the ocean surface, leading to significant optical, biological and geochemical responses. With satellite ocean color remote sensing, e.g., from the Moderate Resolution Imaging Spectroradiometer (MODIS), biological and geochemical responses following a hurricane are revealed. Dramatic changes of sea surface reflectance are observed following a hurricane or a cyclone [Shi and Wang, 2008; Hu and Muller-Karger, 2007]. The increased nutrients on the sea surface after a hurricane can also enhance ocean productivity and trigger phytoplankton blooms [Shi and Wang, 2007; Lin et al., 2003; Walker et al., 2005]. Chlorophyll-a concentrations in the phytoplankton bloom regions show a 20-to-30-fold increase following Hurricane Katrina [Shi and Wang, 2007] and Cyclone Kai-Tak [Lin et al., 2003]. Synoptic increases of the surface chlorophyll-a concentration within the cool wakes of the hurricanes are also observed [Babin et al., 2004]. The post-storm surface chlorophyll-a enhancement usually lasts ∼2 to 4 weeks.

[4] The ocean's physical response to a hurricane is right-biased. This is attributed to the combined effect of the asymmetrical hurricane wind stress field and the near-inertial response of the ocean mixed layer to the asymmetrical surface wind stress [Price, 1981]. However, it is largely unknown whether the ocean's biological response to a hurricane is also right-biased. Few satellite observations have ever addressed the synoptic asymmetry of the ocean's physical and biological responses to a moving hurricane because most hurricanes that have significant environmental impact are not in open-ocean. Thus, the asymmetrical geography, bathymetry, and ocean environments make it difficult to evaluate the asymmetrical features of ocean's responses in large scale with satellite remote sensing observations.

[5] The objective of this paper is to evaluate the asymmetrical features of ocean physical and biological responses and investigate the changes of ocean vertical profiles using Hurricane Earl that moved north over deep water off the U.S. east coast as an example. Advanced Microwave Scanning Radiometer EOS (AMSR-E) SST and MODIS-Aqua ocean color products are analyzed for the synoptic sea surface physical and biological variability before and after the hurricane. AMSR-E measurements produce microwave SST under both cloudy and clear sky conditions [Wentz et al., 2000]. Therefore, AMSR-E data are really useful for continuous observations of the SST for the periods before, during and after the hurricane passage. Chlorophyll-a data are from the NASA MODIS ocean color standard products. It is found that for the study region the shortwave infrared (SWIR) atmospheric correction algorithm [Wang, 2007] is not required for the ocean color data processing. National Centers for Environmental Prediction (NCEP) sea surface wind data and Argo (www.argo.net) float temperature and salinity profile data are also used. Since the Argo float measures the temperature and salinity profiles in the upper 1100 m within a time frame of ∼9 days, it can provide a valuable opportunity to assess the ocean's vertical response to a moving hurricane.

2. The Ocean's Asymmetrical Physical Response to Hurricane Earl

[6] As the fifth named tropical storm of the 2010 Atlantic hurricane season, Hurricane Earl is the first major hurricane in this century that moved northward and threatened the U.S. New England region and eventually made landfall in Nova Scotia of Canada. Figure 1a shows the snapshot of Earl's wind fields at 12:00 UTC, September 3, 2010, with the track of the hurricane outlined in red. The hurricane eye location is marked in red squares with intervals of every 6 hours as shown in Figure 1a. At that time, the hurricane center was at (36.2°N, 73.5°W) with the maximum sustained wind speed reaching ∼139 km/h. The hurricane further weakened and was downgraded to a tropical storm late on September 3. On September 4, Hurricane Earl made its landfall in western Nova Scotia with sustained winds of 110 km/h. It is noted that the mean water depth along the hurricane track as shown in Figure 1a is ∼2600 m.

Figure 1.

(a) Snapshot of Hurricane Earl wind fields at 12:00 UTC on September 3, 2010, with the track of the hurricane outlined in red. The hurricane eye locations are marked by red squares with intervals of every 6 hours from September 2–4, 2010. The locations of Argo float R4900442 profiles before and after the passage of Hurricane Earl are also marked with solid red circle. (b) AMSR-E SST fields before Hurricane Earl on September 1 and (c) immediately after the passage of Hurricane Earl on September 5. (d) The SST difference between immediately post-hurricane on September 5 (Figure 1c) and pre-hurricane on September 1 (Figure 1b). The hurricane track is overlaid in each image. Dash lines parallel to the hurricane track in Figure 1b mark the 100 km bands to the hurricane's left and right sides.

[7] The wind fields of Hurricane Earl were not symmetrical to the left and right sides of the hurricane track from NCEP wind data. High winds over 30 m/s dominated the right side of this hurricane within the distance of 150 km from the hurricane eye (Figure 1a). In contrast, the winds were weaker to the left side of the hurricane track. The maximum sustained winds on the left side were typically ∼25 m/s and confined within 100 km off the hurricane track. NCEP wind fields during the hurricane passage between September 2 and 4, 2010 show that the mean maximum sustained wind speed in the area 100 km east (right) of the track was 25.59 m/s, while the mean maximum sustainable wind speed west (left) of the track was 21.96 m/s.

[8] The ocean's asymmetrical physical response was observed by the AMSR-E SST observations. On September 1, the AMSR-E SST image shows typical SST patterns with the Gulf Stream clearly identified in Figure 1b. In the north, the ocean was featured with low SST over the Georges Bank and Gulf of Maine regions. It is noted that there are no SST retrievals in the coastal regions since microwave SST near coasts is invalid.

[9] Following Hurricane Earl, a broad SST drop can be identified along the entire hurricane track. The strong SST front caused by the Gulf Stream before Hurricane Earl was disrupted as shown in Figure 1c. In the South Atlantic Bight region, the relatively uniform SST field before Hurricane Earl was replaced with a low SST trough along the hurricane. In the Mid Atlantic Bight region, SST dropped more than 1°C east of Gulf Stream. In the north, low SST expanded further offshore over the Georges Bank and Gulf of Maine regions. The SST restored two weeks after the passage of Hurricane Earl in the south and the Mid-Atlantic Bight region (results not shown). But the SST along the Hurricane Earl track in the Mid-Atlantic Bight and the Georges Bank regions were not restored to their pre-hurricane values due to the seasonal cooling.

[10] The SST difference between post- and pre-hurricane shows that the SST drop is asymmetrical (Figure 1d). A band of enhanced SST drop is clearly located to the right of the track of the hurricane. As marked in Figure 1b, two banded regions with 100 km width off the hurricane track on both sides are chosen to quantify the asymmetrical feature of the ocean's physical response to Hurricane Earl. The average depths of the left and the right bands are ∼2400 m and ∼2800 m, respectively. Each band covered an area of ∼2 × 105 km2. Table 1 shows the statistics of SST changes following Hurricane Earl. Before Hurricane Earl, the SSTs on the left and right were almost the same. After Hurricane Earl, SST on the right dropped from 25.84°C to 23.99°C, while the SST change on the left was from 25.85°C to 24.62°C. The pre- and post-hurricane SST difference on the right is 1.85°C in comparison to 1.33°C on the left side. After two weeks, the mean SSTs on the right and left side were 23.65°C and 23.96°C, respectively. The SST gap between the two sides dropped from 0.63°C on September 5 to 0.31°C on September 15. The reduced SST difference between the left and right suggests that the band of low SST on the right largely dissipated 10 days after Hurricane Earl's passage.

Table 1. SST Changes on the Left and Right (100 km Off the Track) Along Earl's Track
RegionPre-Hurricane SST (9/1/2010)Post-Hurricane SST (9/5/2010)SST Change (9/5 Post to 9/1 Pre-Hurricane)Post-Hurricane SST (9/15/2010)
Left Band25.85°C24.62°C−1.23°C23.96°C
Right Band25.84°C23.99°C−1.85°C23.65°C

3. The Ocean's Asymmetrical Biological Response to Hurricane Earl

[11] Asymmetrical biological response along Earl's track was also observed with MODIS-Aqua chlorophyll-a observations. As a representative of the biological activity, chlorophyll-a concentration is used to characterize the asymmetrical biological response after Hurricane Earl. Due to frequent cloud coverage in this region, eight-day composite images of chlorophyll-a concentration are used to evaluate the biological activity following Hurricane Earl. Similar to SST evaluation, two bands of 100 km across the hurricane track are employed to further quantify the asymmetrical features of the hurricane-driven biological responses.

[12] Figure 2a shows the typical chlorophyll-a concentration pattern in the study region. A significant chlorophyll-a front separated the eutrophic coastal waters and oligotrophic open ocean waters. After Hurricane Earl, a broad increase of chlorophyll-a concentration was observed along the hurricane track (Figure 2b). Specifically, in the South Atlantic Bight, chlorophyll-a concentration showed modest gains. In the Mid-Atlantic Bight region, a filament of enhanced chlorophyll-a as marked in Figure 2b was also observed to the right of Hurricane Earl's track. In the Georges Bank and Gulf of Maine regions, chlorophyll-a concentration was enhanced significantly following the passage of Hurricane Earl.

Figure 2.

Chlorophyll-a concentration distributions derived from MODIS-Aqua observations (a) before Hurricane Earl between August 24 and September 1, (b) after Hurricane Earl between September 4 and September 12, (c) after Hurricane Earl between September 13 and September 21, and (d) chlorophyll-a ratio between post-hurricane chlorophyll-a (September 4–12) and pre-hurricane chlorophyll-a (August 24–September 1). The hurricane track is overlaid in each image. Dash lines parallel to the hurricane track in Figure 2a mark the 100 km bands to the hurricane's left and right sides. In Figure 2b, regions of notable chlorophyll-a elevation are outlined in white.

[13] Figure 2c shows the chlorophyll-a concentration fields on September 13–21. Even though the modest enhancement of chlorophyll-a still existed in the South Atlantic Bight region, the chlorophyll-a concentration elevation in the Mid-Atlantic Bight and further north was not as significant as shown in Figure 2b. Particularly, the stretch of chlorophyll-a enhancement filaments immediately after Hurricane Earl's passage could not be identified in the MODIS-Aqua chlorophyll-a image.

[14] Figure 2d shows the ratio of chlorophyll-a concentration after (Figure 2b) and before (Figure 2a) the passage of Hurricane Earl. For Earl's entire track of nearly 2000 km along the U.S. East Coast, the enhancement of biological activity as represented by chlorophyll-a concentration was mostly right-biased. This is especially true for the biological response in the Mid-Atlantic Bight and Georges Bank regions. The pattern of chlorophyll-a enhancement in Figure 2d does not exactly match the spatial pattern of SST drop (Figure 1d). This suggests that the biological response to a hurricane is a more complicated process, which relates to various processes such as the physical, optical and geochemical. It is noted that in similar periods (between two weeks), MODIS-Aqua 2002–2008 chlorophyll-a data show that the chlorophyll-a natural variability in the Hurricane Earl-track region is about 7–8%. Table 2 further quantifies the asymmetrical feature of the ocean's biological response to Hurricane Earl. Indeed, the enhancement of chlorophyll-a concentration occurred on both sides. Since the left band is further closer to the coast, its mean chlorophyll-a concentration was larger than those to the right side before the hurricane's passage. Chlorophyll-a concentration jumped from 0.15 to 0.31 mg m−3 in the right band (by a factor of 2.04), while its value increased from 0.25 to 0.34 mg m−3 for the left band (only by a factor of 1.33), demonstrating that the biological response on the hurricane's right side is more intense than the response on the left side, which is similar to the asymmetrical feature of the ocean's physical response to Hurricane Earl.

Table 2. Chlorophyll-a Concentration Changes on the Left and Right (100 km Off the Track) Along Earl's Track
RegionPre-Hurricane Chl-aa (8/24–9/1/2010)Post-Hurricane Chl-aa (9/4–09/12/2010)Chl-a Ratio (Post-Over Pre-Hurricane)Post-Hurricane Chl-aa (9/13–9/21/2010)
  • a

    Unit in mg/m3.

Left Band0.2510.3351.330.325
Right Band0.1520.3102.040.220

4. The Ocean's Physical Response in Vertical to Hurricane Earl

[15] When Hurricane Earl passed off the U.S. East Coast, Argo float R4900442 was located very close to Hurricane Earl's track. The locations of Argo float R4900442 are marked in Figure 1a before and after Hurricane Earl. The float measured the temperature and salinity profiles (profile ID: R4900422_248) at 6:00 UTC on August 29 at a location of (38.663°N, 70.567°W). Ten days later, it made another measurement (profile ID: R4900422_249) at 9:00 UTC on September 8 at a location of (39.162°N, 70.680°W). The distance between these two measurements was ∼56 km.

[16] Figure 3a shows the comparison of the temperature profiles before and after the passage of Hurricane Earl. Similar to the AMSR-E SST observations, the temperature in the mixed layer of the upper 20 meters showed a ∼2–3°C drop following Hurricane Earl. Thermocline was lifted from ∼20–70 m depths to ∼20–40 m depths. This suggests that extensive vertical mixing and upwelling occur in the upper 70 m water column, this observation is consistent with the modeling study with Hurricane Katrina [Liu et al., 2009]. The profile of the temperature before and after Hurricane Earl also shows that the hurricane-driven cooling dominated the upper 70 m with the highest temperature drop at 40 m depth. A slight temperature drop was also seen below the thermocline following Hurricane Earl. This indicates that the vertical mixing indeed can occur in the deep layer, but it was much reduced in comparison to the vertical mixing in the mixed layer and the thermocline.

Figure 3.

Profiles in the upper 200 m water column measured by Argo float R4900442 for (a) pre- and post-hurricane ocean temperature and (b) pre- and post-hurricane ocean salinity. Profile 248 was measured at a location of (38.663°N, 70.567°W) before Earl at 6:00 UTC on August 29, and profile 249 was measured at a location of (39.162°N. 70.680°W) after Earl at 9:00 UTC on September 8. These two locations are indicated in Figure 1a.

[17] Due to strong hurricane-driven mixing, significant elevation of salinity was observed with the Argo salinity profile measurements. In the surface mixed layer, salinity increased from 34.74 psu before Hurricane Earl to 35.35 psu after (Figure 3b) due to hurricane-driven vertical mixing. The depth of the maximum salinity was lifted from 90 m to 70 m. In addition, unlike the change of temperature profile, salinity change tapered with the depth. At a 40 m depth, the salinity after the hurricane was actually slightly lower than before the hurricane. The discrepancy of the temperature and salinity changes is attributed to the different profile shapes of these two parameters. In the deep layer, the slight change of the salinity also suggests that the vertical response was significantly reduced at this depth.

5. Discussions and Summary

[18] In this study, we demonstrate and quantify the asymmetrical features of physical and biological responses to a hurricane with MODIS observations. Argo float in situ temperature and salinity profiles are used to evaluate the vertical response to Hurricane Earl. In the North Atlantic region, most of previous studies with regards to ocean's response to a hurricane were focused in the coastal regions. The complexities of the coastal ocean processes, such as phytoplankton blooms, sediment re-suspension, river plumes and frequent ocean eddies, often overshadow the asymmetrical features of the ocean's responses to a hurricane, especially the optical and biological. In contrast, Hurricane Earl passed a vast region of open ocean farther off the U.S. East Coast before landing. The ocean's relatively uniform nature across the hurricane at the sea surface and long distance track farther offshore provide a unique opportunity to assess the difference of the ocean's response across the hurricane track. The SST difference (Figure 1d and Table 1) and chlorophyll-a ratio (Figure 2d and Table 2) following Hurricane Earl effectively demonstrated stronger response to the right of the hurricane's track in spite of the possible discrepancies in the initial ocean conditions to both sides.

[19] Babin et al. [2004] reported that the ocean generally shows 20%–80% increase of chlorophyll-a concentration and 0.5–2.5°C drop of SST after analyzing 13 hurricanes passing through the Sargasso Sea from 1998 to 2001. The increase of the post-storm chlorophyll-a concentration lasted approximately two to three weeks. In this study, the chlorophyll-a concentration increase, SST drop, and the duration of the chlorophyll-a enhancements following Hurricane Earl fell just within those ranges with notable disparities on the hurricane's left and right sides. Argo float in situ data show the shoaling of the mixed layer after Hurricane Earl. This is different from the deepening mixed layer depth of Babin et al. [2004]. Even though color-dissolved organic matter (CDOM) can also be elevated in the North Atlantic region following a hurricane [Hoge and Lyon, 2002], the ocean color spectrum in the deep blue (412 nm) and blue (443 nm) wavelengths and the fluorescence line height (FLH) over the regions of enhanced chlorophyll-a after the passage of Hurricane Earl provide evidences that the elevation of the chlorophyll-a is indeed caused by the increase of phytoplankton concentration. On the other hand, since the chlorophyll-a concentration changes after a hurricane passage is well correlated to the increase of the nutrient concentrations [Shi and Wang, 2007; Liu et al., 2009; Babin et al., 2004], the asymmetrical features of the SST and the chlorophyll-a concentration along Hurricane Earl's track also indicate that the response from the marine nutrient fields, such as nitrate, phosphate and silicate, might also be asymmetrical across Earl's track. But, for verifying this further studies are needed.

Acknowledgments

[20] This research was supported by NASA and NOAA funding and grants. MODIS chlorophyll-a data were obtained from the NASA/GSFC ocean color website. AMSR-E data are produced by Remote Sensing Systems and sponsored by the NASA Earth Science REASoN DISCOVER Project and the AMSR-E Science Team. Argo data were collected and made freely available by the International Argo Project and the national programs that contribute to it (http://www.argo.ucsd.edu, http://argo.jcommops.org). Argo is a pilot program of the Global Ocean Observing System. We thank two anonymous reviewers for their useful comments. The views, opinions and findings contained in this paper are those of the authors and should not be construed as an official NOAA or U.S. Government position, policy or decision.

[21] P. G. Strutton thanks two anonymous reviewers.