Revisit Precipitation Units

[rmartz]

#Because precipitation is measured in the raw data using obscure units of rate, kg/m^2/s, in our indicator response the values are also units of rate... we convert kg/m^2 to linear distance by using the density of water as 1000 kg/m^3, but keep the time denominator as return values such as in/day.

Currently there are only three precipitation indicators (Two after #162):

• YearlyTotalPrecipitation
• MonthlyTotalPrecipitation
• DailyPrecipitation

Because precipitation is measured internally as average over the entire day, and each of these indicators are a total of varying levels of granularity (Total precipitation over a year, month, and day, respectively), I think it makes sense to use units that do not include a time component. Whether the user requests precipitation to be measured in mass or linear distance, the time component indicates the number as a rate, while the indicators suggest a total.

Rate would be desired for an indicator like HeaviestPrecipitation, which could be measured in in/hr, for instance, but the total precipitation over an aggregation period would be measured in kg/m^2, in or mm (For example in WeatherDB)

[rmartz]

This has been changed by #162 so closing here