The RainWise MKIII and WS-2000 specifications are quoted below.
These are factory claimed specifications that the unit presumably met when shipped. Confirmation of the validity
of these specifications can only be accomplished in a laboratory environment where the meteorological
variables are precisely set and carefully applied to the weather station under test. Once the station is
installed, it is difficult to evaluate the precision of its measurements. There are, however, methods available
to test the authenticity of a given stations' measurements.
One of them involves using independent hardware of known precision to make side-by-side comparisons with the installed station. At various times, I have made wet and dry bulb temperature and relative humidity comparisons with the MKIII. Both a sling psychrometer and a solid state temperature/humidity recorder have been used. A good bit of care must be made in order not to corrupt the test. For example, the test must be made on a cloudy day so that sunlight will not accidently heat the test sensors. Occasionally, the results were inconclusive but usually the test showed that the MKIII was probably maintaining its original precision.
The second method is done on a daily basis by NOAA. Data from this station is submitted to NOAA once every 15 minutes. Since the data will be used by NOAA for preparing forecasts and a myriad of other scientific tasks and projects, it must be examined carefully to determine if it is trustworthy. NOAA compares my data with similar stations located nearby. Visit the Data Quality page to see this comparison. In my case, however, the nearest station reporting data is more than 10 miles away; it is well known that all meteorological parameters can show wide variability over this distance scale. However, NOAA's computer model takes into account some of the temporal and spatial differences between stations. Some days, the atmosphere is nearly homogenious and the meteorological parameters are "flat" over wide regions. This makes it easier to spot data from a station that is biased. Other situations like station siting will produce data that is corrupt and these are also easy to detect. But, in a carefully installed station, the NOAA comparisons might only reveal a sensor failure at best. Data "scatter" plots are also produced. These are diagrams that show the standard deviation plotted against the mean error for each parameter at each station over a number of days (usually14 days). The scatter plots show whether the station is inside or outside of quality control boundaries. Because the statistics used to obtain these plots is credible their results can be used as a basis for making adjustments to the sensors of a station but only after a long series of mean errors are observed.