Study notes on infrared (IR) sea surface temperature (SST) imaging and ocean color Sea surface temperature Applications Boundary condition to weather and climate forecasting and analysis Visualization of ocean circulation Upwelling, eddies, fronts, geostrophic turbulence, ENSO Data assimilation in ocean forecasting models Climate variability analysis Tech IR instruments: AVHRR, MODIS, ATSR … Satellite platforms: NOAA-n, METOP, Terra, Aqua, ERS, Envisat, GOES, METEOSAT, … Resolution: ~300 m for ATSR, ~1 km for AVHRR, ~5 km for sensors of geostationary satellites like current generation of GOES IR channels for SST Wavelengths of emitted radiation depend on black-body temperature – earth emits mostly in the IR (Wien’s Law). IR radiation reaching a satellite depends on emission from SST, but also by the intervening atmosphere (gases, water vapor, clouds, and aerosols) Transmittance of atmosphere in IR varies significantly with wavelength because of absorption by various gases – there are a few “windows” (small ranges of wavelength) in which there is high transmittance, e.g. ~3.5 micrometers, 10.8 micrometers, 12 micrometers. IR imagers use a set of channels (the wavelengths detected by the instrument) which are windows in transmittance, and also have a strong signal (i.e. near the center of the black-body spectrum of the emitted radiation at 10 and 12 micrometers) or sensitivity totemperature (i.e. where the emittance spectrum changes rapidly with SST at 3.5 micrometers). Other things you should know 3.5 micrometers is short enough to include some reflected sunlight, so becomes an unreliable channel during daytime. Absorption coefficient (an inverse length scale) of IR in water is ~ 10-6 m so emittance is from surface 1 mircometer only. IR shows the skin temperature, but is typically reported as a bulk temperature because of the way the algorithm is calibrated against ~1 m depth in situ observations from drifting buoys, moorings, and ships. A “split-window” algorithm uses IR observations in multiple channels to account for absorption of IR in the atmosphere. Clouds obscure IR so need to be identified in data: Cloud tops are cold – temperatures less that zero are not water SST tends to vary modestly on length scales of a few km, so large consistencies between adjacent pixels may indicate the presence of patchy clouds. In a multi-pixel view, taking the warmest (or, say, 95-percentile value) temperature is another strategy to excluding values biased cold by the presence of sub km scale cloud. Volcanic aerosols can lower apparent temperature by as much as 0.5 to 2 degrees. Study notes on CODAR HF radio frequencies roughly 5 to 45 MHz with range of operation from ~200 km to ~20 km, respectively Bragg scattering causes signal to be dominated by a narrow range of ocean surface wave with wavelengths in the range 6 m (for 25 MHz CODAR) to 30 m (for 5 MHz CODAR) Ocean wave theory gives the speed at which ocean waves travel as a function of their wavelength. Because of the Bragg scattering phenomenon, we know which ocean waves are being imaged and how fast they travel is there is no ocean current. The speed of the waves Doppler shifts the radio reflection. Spectrum processing reveals two peaks almost centered on the radio carrier frequency: a peak Doppler shifted to higher frequency corresponds to the ocean waves approaching the receiver. There is a similar peak shifted to lower frequency for ocean waves traveling away from the receive antenna.If the two peaks are not exactly centered on the carrier frequency, there is an added Doppler shift due to an ocean current flowing toward or away from the antenna. With two or more receive antennas, components of the ocean current in multiple directions can be estimated and the total vector velocity computed. The data are geospatially located with two aspects to the system design: Azimuth, or angle in the horizontal around the receive antenna, is determined by having multiple loops with different directional gain. With a known directional gain, the azimuth of the target can be determined. Range, or distance offshore from the receive antenna, is determined by modulating the transmitted signal with a swept-frequency carrier signal and demodulating it properly in the receiver. The time delay between transmit and receive is then converted to a large-scale frequency shift in the echo signal. The first step in digital spectral analysis of the signal extracts the range or distance to the sea-surface Bragg scatterers, and sorts it into range bins. These bins are typically 1 to 12 km in width. The Doppler processing is applied to the signals that have been separated into each range bin. Data are typically processed for 4 to 20 minutes to reduce noise to a level capable of resolving the Doppler shift peaks in the spectrum. Then the ocean current estimates are averaged for 1 to 3 hours to produce the final radial current estimate for each range/azimuth bin. The ocean current estimate is for the flow in the top 0.5 (for 48 MHz systems) to 2 m (for 5 MHz systems) of the water. Higher frequency systems have less range but better spatial resolution. High frequency 48 MHz systems tend to be used near port entrances where detailed current patterns aid navigation. Range is about 20 km. Resolution is 0.25 to 1 km. Lower frequency 5 MHz systems are used in networks along coasts to image currents well offshore. Range is about 200 km. Resolution is 6–12 km. Ocean Color Applications Observation of ocean phytoplankton biomass, dissolved organic matter, particulate carbon, suspended sediment. Derived products include net primary productivity. What makes the ocean colored?Phytoplankton pigment, colored dissolved organic matter (CDOM), organic detritus, silt, clay Case 1 waters: refers to open ocean where color is dominated by phytoplankton and their detritus Case 2 waters: refers to coastal regions where CDOM and suspended inorganic silt/clay are also optically significant Ocean color sensors view reflected light. The ocean does not emit visible colors. Inherent Optical Properties (IOP) such as absorption due to pigments are independent of the light source. Transport conservation equations can be derived for IOP because the mix and diffuse according to their concentration. Apparent Optical Properties (AOP) are how the reflectance varies depending on illumination angle and