Woods Hole Coastal and Marine Science Center


next up previous contents
Next: Velocity Time Series Up: Model/Data Comparison Previous: Model/Data Comparison

Temperature, Salinity and Density

The comparison between modeled and observed temperature, salinity and density at most sites shows that the model generally does a good job at reproducing seasonal cycles and top to bottom density differences (Figures 3.1-3.8). There are some clear differences however. While most of surface salinity variations during the spring and summer are well represented (Figure 3.1), the salinity drop associated with the large fall 1990 discharge appears in the model about three weeks too late, and during spring of 1991 the surface salinities in the model are 0.5-1.0 psu too salty. The lower layer salinities (Figures 3.2 and 3.3) during this time period are also 0.5--1.0 psu too salty, so the vertical density structure is minimally affected. It does suggest, however, that if a better match of the salinity magnitude was desired it would be necessary to improve upon the climatological salinity boundary conditions for the western Gulf of Maine model. These boundary conditions specified a fixed seasonal cycle, assuming that all interannual variability was due to local river discharge in the western Gulf of Maine. It is reasonable to expect that interannual variability in discharge from the St. Lawrence river and variations in the Gulf of Maine circulation would also be responsible for interannual salinity fluctuations in the water entering Massachusetts Bay.

 Plots of model/data comparison of near surface salinity

 Plots of model/data comparison of lower salinity

 Plots of model/data comparison of lower salinity

  Plots of model/data comparison of near-surface temperature
Figure 3.4: Model/data comparison of near-surface temperature at stations BB, BS, SC and U6.

 Plots of model/data comparison of near-surface temperature

 Plots of model/data comparison of lower temperature

 Plots of model/data comparison of lower temperature

 Plots of model/data comparison of vertical density

The modeled temperatures and salinities appear to track some of the observed shorter period fluctuations in addition to the general seasonal cycle. The coherence between modeled and observed water properties is generally significant at 95% confidence over periods of 2--20 days (Figures 3.9 and 3.10). The temperature coherence decreases somewhat with increasing frequency, but the salinity coherence doesn't show any obvious dependence on frequency.

 Plots of data coherence

 Plots of temperature coherence

Model fluctuations in salinity are comparable in strength to observations (Figure 3.9) except over Stellwagen Basin (U6), where the modeled salinity fluctuations are less that half the strength of the observations. The strong observed fluctuations of near-surface salinity at U6 (Figure 3.1) are associated with intrusions of low-salinity water from the Gulf of Maine (Geyer et al. 1992), and it appears that these intrusions are either under-represented or are somewhat smeared out by the model.

Model fluctuations in temperature are comparable in strength to observations (Figure 3.10) except at Scituate (SC) and Broad Sound (BS), where the modeled temperature fluctuations are about 50% stronger than the observations. The strong modeled fluctuations of near-surface temperature at these near-shore locations are associated with upwelling events, so one explanation is that the model surface layer responds a bit too strongly to upwelling winds.


next up previous contents
Next: Velocity Time Series Up: Model/Data Comparison Previous: Model/Data Comparison

Skip Navigation

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USAGov logo U.S. Department of the Interior | U.S. Geological Survey

URL: woodshole.er.usgs.gov/operations/modeling/mbayopen/node23.html
Page Contact Information: WHSC Webmaster
This page last modified on Monday, 24-Nov-2014 13:04:14 EST