The temporal variation of the calibrated irradiances with time are shown in the figure below.

The mooring deployment periods are bracketed by the vertical black bars.  The mooring deployment numbers  (1-7) are shown along the top of the figure.  Note that data from before day 60 for the Ed670 and Ed683 channels were removed, as were data from between day 260 and day 332 for all channels (see data processing).   The lack of data from the third deployment was caused by corrosion of the data logger.  A similar problem caused a lack of data after day 768 in the fifth deployment.  No data are available from deployment seven because the sensor was lost.

There are several cautionary notes that should be made regarding this data set:

• Three different irradiance sensor heads have been used.  The first, used in 1997 (blue symbols) had an Ed(510) channel.  The second, used in 1998 and 1999 (red symbols) had an Ed(700) channel instead.   The third, used in late 1999 (green symbols) had an Ed(510) channel.  The resolution of these instruments also differed - the first had a range of 0 - 4095 counts, the second had a range of 0 - 65535 counts, and the third had a range of 32767-65535 counts.
• Note the abrupt change in irradiance values for the Ed670 and Ed683 channels from the end of the 1st deployment to the beginning of the second deployment.  At the same time, the lower wavelength channels appeared to have remained relatively constant.  The reason for the abrupt change is unknown, hence the Ed670 and Ed683 data from 1997 should be treated with caution.
• No calibration was made between the fourth and fifth deployments, however the sensors were cleaned.  Note the abrupt change in values for the longer wavelength channels between the two deployments.   This may have been caused by biofouling of the sensor towards the end of the fourth deployment.
• The levelling-off of irradiance values observed in the shorter wavelength channels (especially Ed443 and Ed490) of the first two deployments is partly caused by saturation of these irradiance channels at high irradiance levels.   This problem was amended in 1998 with the deployment of the new instrument.  The saturation values for each of the channels during each deployment are listed below:

Webpage  by Jasmine S. Bartlett, Oregon State University.