As hard as we try, bugs do occasionally occur in our software. Please accept our sincerest apologies!
CASPR v3.2.3, September 7, 2001: A bug in the 'fill_clear' procedure will result in incorrect interpolation of clear sky values to cloudy sky pixels, which will therefore affect cloudy sky parameters. The effect of this bug is expected to be small overall, but can be significant (e.g., a few degrees in surface temperature) in coastal areas.
CASPR v3.2.1, May 2, 2001: A bug in the time series cloud mask procedure that will reduce the number of clear pixels used in the clear sky statistics. The effect of this bug is expected to be small.
CASPR v3.2, February 2, 2001: A bug that results in radiative fluxes not being computed for clear areas. Instead, fluxes for clear pixels will get a value of MISSING. This is certainly something that will be noticed when viewing the results, but may not be noticed if you are only looking at area averages. This may also be an issue with earlier versions (3.0 and 3.1) as well.
FluxNet v2.1 and CASPR v2.4.x, April 6, 1999: A bug was found in FluxNet that affected both FluxNet and the radiative flux calculation in CASPR. The bug affects radiative fluxes (only) in CASPR under all conditions. It affects all output in FluxNet if cloud water content is anything other than 0.2 g m-3 for water clouds or 0.05 g m-3 for ice clouds.
The bug has been fixed in CASPR version 2.5, which is now available. FluxNet can be used as is if cloud water content is input as specified in the updated manual. A new version of FluxNet is under development, so I'm not going to bother "fixing" the current one.
Streamer is NOT affected by this bug and continues to be a stable and reliable tool.