|
University of California, Santa Barbara |
|
|
University of California, Santa Barbara |
Introduction
Over the past year we have continued development of a network of high frequency (HF) radars along the northern coastline of the Santa Barbara Channel and the central coast north of Pt. Conception, California. The radars are commercial systems called Coastal Ocean Dynamics Radars (CODAR’s) manufactured by CODAR Ocean Sensors, Ltd. of Los Altos, CA. The network of HF radars is used for mapping ocean surface currents from coastal sites out to a range of about 40 km. They operate by transmitting radar waves over the sea surface and then observing the characteristics of the back-scattered radar waves received at the radar sites. Doppler shifts in the received signals can be used to estimate surface current velocities around the radar sites. By combining signals from all sites, maps of surface currents can be generated every hour.
Major goals for the project are to:
- evaluate HF radars for observing evolving patterns of surface currents
- observe and describe evolving surface currents in the Santa Barbara Channel and Santa Maria Basin
- study the relationship between evolving surface current patterns and spatial variability in various biological populations. These populations include intertidal organisms and pelagic juvenile fishes
HF radar array along the South-central California Coast
A major accomplishment during the past year has been the establishment of two HF radars at sites on Vandenberg Air Force Base (VAFB). This process took two years and innumerable phone calls, faxes, and emails, in addition to many changes in our project contact personnel at VAFB. The two new sites are at Pt. Arguello and Fallback-22, a site just south of Pt. Sal. The equipment for these sites is on loan to our group from the NOAA Environmental Testing Laboratory (NOAA/ETL) in Boulder, CO. Our colleague there, Jack Harlan, has been very helpful in arranging for the loan and consulting with our group on HF radar operations and analysis. Our group has greatly benefited greatly from our interactions with NOAA/ETL.
We now have HF radar systems operating at the following locations (from south to north):
site 1: Coal Oil Point – on the UCSB campus
site 2: Refugio – at the operations center for the Channel Coast Ranger District
site 3: Pt. Conception – at the lighthouse
site 4: Pt. Arguello – on VAFB
site 5: Fallback-22 – on VAFB
We are able to provide real-time data from two of the sites, Coal Oil Point and Refugio. This has proven extremely valuable in conducting real-time experiments in the western Santa Barbara Channel in which we direct shipboard sampling based on observed surface current patterns. There are problems with the phone line at Pt. Conception so we cannot remotely retrieve data from this site at present. Diagnosis and repair of this phone line has proven difficult due to our lack of access to this site located on the Bixby Ranch (see below). We have been unable to get Communication Services at VAFB to install phone lines at the Pt. Arguello and Fallback-22 sites. However, we are exploring the possibility of using cell modems or radio links to retrieve data from the latter three sites.
Evaluation of HF radar performance
A major focus of our research continues to be evaluating HF radars as instruments for the measurement of surface currents. Our approach is to make two fundamental comparisons.
First we compare time series of currents at selected locations from the HF radars with in situ currents measured from moored current meters located in our coverage area. Second we compare currents measurements from a single HF radar site with predictions of those measurements based on data from the other sites. This provides a check on the consistency of the HF radar technique. We have two sources of in situ current data: the extensive array of moorings operated by the Center for Coastal Studies at the Scripps Institution of Oceanography (CCS/SIO); and a mooring operated by our research group at UCSB located about 18 km south of the UCSB campus.
We compare radar and current meter measurements by producing time series of HF radar-measured currents at several locations surrounding available current meter moorings. These are compared statistically with simultaneous current time series obtained by the current meters, located typically at 5 m depth. The normalized covariance between radar and the current meter time series, called r2 or “r-squared,” is computed for each grid location. We find that maximum r-squared values are in the range .55 to .72, indicating that roughly half to ¾ of the variance in the radar time series of surface currents can be explained by the currents measured at 5 m depth by the current meters. Differences in the time series may result from several factors including different spatial sampling, current shear near the surface, wave effects, and measurement noise (Graber, et al., 1997). An interesting result from this analysis is that there is usually a bearing error in the radar measurements, typically in the range 5 – 15 degrees. We speculate that this results from uncertainties in the antenna patterns for each radar system. We are working with colleagues in the HF radar community and at CODAR Ocean Sensors to understand and correct for this error. The effects of this error appear to be relatively small. Results from this analysis were presented at the Fall Meeting of the American Geophysical Union (Emery and Washburn, 1998).
Another comparison we make is between time series from different radar systems. In this procedure we find an area of the ocean’s surface that is simultaneously observed by three radar systems. Current vectors are first computed between two sites. Then these are used to predict the radial current vectors observed by the third site. Radial currents are those observed by a single site; at least two sites are required to estimate the total current vector at a point on the ocean’s surface. This work is on-going, but preliminary results indicate that about 50-60% of the variance in radial currents from one site can be explained by those estimated from surrounding sites. We are working to refine this comparison technique and understand measurement differences.
Surface current patterns on the South-central coast
Recent studies have shown that the circulation in the Santa Barbara Channel tends to be cyclonic, or counter-clockwise, much of the year (Dever et al., 1998, Hendershott and Winant, 1996, Harms and Winant, 1998). The cyclonic circulation results from a combination of local and remote forcing processes over a wide range of spatial scales. Specifically, it results from the competition between wind forcing to drive eastward flow in the Channel and sea level difference outside the Channel to drive westward flow.
We have been exploiting the ability of HF radars to observe on scales of a few km to map the evolving circulation in the Channel. Maps of vorticity, a measure of the rotation rate of water parcels, clearly shows a strong tendency for cyclonic flow, at least for 1998. Often the flow is organized into a large counter-clockwise eddy that spans most of the Channel. The location of the eddy usually coincides with the Santa Barbara Basin. Because of this we are investigating the hypothesis that the eddy is steered by the bathymetry.
To date of our results are based on data from 3 radar sites operating in the Santa Barbara Channel. We are now processing data from our two sites north of Pt. Conception so that the circulation patterns between UCSB and Pt. Sal can be observed. Our preliminary observations suggest than an anti-cyclonic (clockwise) eddy exists in the Santa Maria basin, consistent with observations made by CCS/SIO based on drifter and moored current studies. The characteristics of this eddy will be an important focus of our future research.
Interdisciplinary studies
A major focus of this research is to understand what role coastal circulation processes play in the recruitment of organisms such as barnacles, crabs, and other invertebrates into intertidal habitats. We are using the HF radars to observe patterns of coastal currents over an area in which intensive sampling of intertidal recruitment is taking place. This effort has been augmented with funding from the David and Lucille Packard Foundation. Funding from the Packard Foundation will permit us to increase the scope of the project to include sites on the Channel Islands and to examine recruitment of various fish species into sub-tidal habitats. We have recently deployed an array of moorings at 6 sites within the coverage of the CODAR array. These moorings include a variety of larval recruitment collectors. 7 additional moorings will soon be deployed around Santa Cruz Island. These moorings should yield time series of larval recruitment for a variety of invertebrate species that can be compared to circulation events recorded by the radars. Our ongoing measurements of settlement patterns have used collectors suspended from piers. Although these samples have identified abrupt changes in settlement dynamics at two locations in the vicinity of Pt. Conception, the number and location of piers is not ideal for this sampling. The new moorings will allow us to sample larval recruitment at key locations spanning the biogeographic transition at Pt. Conception.
Another interdisciplinary study arising from this project is an examination of the distribution of juvenile fishes around the Santa Barbara Channel in relation to a strong cyclonic eddy that often occurs over the Santa Barbara Basin. Data from June, 1998 revealed much higher abundance of juvenile fishes and late-stage larval fishes in the eddy compared with surrounding waters. We are conducting a second survey of juvenile fishes in the same region at the time of this writing (June, 1999).
Collaborations
We are collaborating with physical oceanographers at CCS/SIO in our investigation of circulation processes in the Santa Barbara Channel and Santa Maria Basin. Our work compliments their efforts since we can observe processes with higher spatial resolution. We benefit from the data collected by their extensive array of current meter moorings in our coverage area.
Our collaboration with Mary Nishimoto and Milton Love of UCSB’s Marine Science Institute has resulted in the investigation of juvenile fishes discussed above. We continue to work with Jack Harlan of NOAA/ETL on improving the operation of HF radars and interpreting the resulting data. Over the past year we have provided personnel of the National Park Service and the Channel Islands National Marine Sanctuary examples of flow patterns in the Channel. Among other issues, they are interested in how current patterns might result in the transport of organisms and pollutants between the mainland coast and the Channel Islands.
Action requested from MMS
A serious impediment to our research is the continuing problem with access across the Bixby Ranch surrounding Pt. Conception. In September, 1998 we were denied access to Bixby Ranch property, despite the fact we had been crossing their property for the previous year without incident. So far as we can determine, we were denied access through no fault of our own. We need help from MMS to gain access to our site at Pt. Conception. The Pt. Conception area is of major scientific interest and it is essential that some sort of accommodation be reached with the Bixby Ranch Company soon. We will do whatever we can to facilitate this.
References
Dever, E.P., M.C.
Henderschott, and C.D. Winant (1998), “Statistical aspects of surface drifters
observations of circulation in the Santa Barbara Channel, J. Geophys. Res.,
103, C11, 24,781-24,797.
Graber, H.C., B.K. Haus,
R.D. Chapman, and L.K. Shay (1997), HF radar comparisons with moored estimates
of current speed and direction: expected differences and implications, J.
Geophys. Res., 102, C8, 18,749-18,760.
Harms, S., and C.D. Winant, (1998), Characteristic patterns of circulation in the Santa Barbara Channel, J. Geophys. Res., 103, C2, 3041-3065.
Hendershott, M.C., and C.D.
Winant, (1996), The circulation in the Santa Barbara Channel, Oceanography,
9(2), 114-121.