Potential improvements for more efficient watering?

[djrobx]

California is currently in a very bad drought.   While the EtherRain has been helpful in making scheduling easy, I feel its weather-specific features to optimize watering is very lacking.   At best I can control the amount of water on a month to month basis, and I'm limited to a single upper temperature for "hot days" to trigger additional watering.   

Our local newspaper had an article about coming water restrictions, and I was surprised to see the first comment was an insightful post about using CIMIS (California Irrigation Management Information System) to calculate the amount of water required to replenish his lawn based on the evaporation from previous day(s).      It has a RESTFUL web API that will report the amount of evaporation (in inches).     You can use this to calculate the amount of inches of water you need to replace when you irrigate.   You can place a cup on the lawn to figure out how many inches per minute your irrigation system outputs.

It seems like this is exactly what's missing.  Using weather underground or yahoo to forecast precipitation is great, but some intelligence around how much watering is actually needed could result in additional water savings.    We have mild weeks, hot dry weeks, and hot humid weeks.    A smart controller should adapt. 
   

[Jim]

Hello djrobx,
The type of irrigation that you are referring to is called Evapotranspiration (ET).  The system you are referring to that is run by CIMIS in California provides the needed information to run Evapotranspiration type cycles. 

While the ET based systems are theoretically the most optimized form of irrigation, in practice they are difficult to implement and don't always provide optimization.  Example:  ET was designed for agriculture to provide growers with irrigation information for optimizing their crop yield.  The formula takes weather elements and crop type into account.  The method is based on assumptions that hold for agricultural irrigation environments but don't hold for most residential irrigation installations.  That is:  1.  the "crop" is constant;  2.  The irrigation area is uniform.   In residential irrigation, there are many "crops" - lawn, garden, trees, shrubs.  In residential irrigation the irrigation area is not uniform:  tree and building shadows, different soils.   

The basic assumptions upon which ET is based for agriculture are not valid for residential - in most cases and so this form of irrigation isn't as optimal as some would have you believe.  To make ET work you need additional information:  inches of water applied per minute or hour; and a good weather station with a rain gauge.   Also, the CIMIS system doesn't account for microclimates so you have to make adjustments to the formula.

In residential use ET is complicated, requires additional information that many installations don't have, and requires adjustments.   Our monthly setback function is an ET estimation.  It is a simple number system that most can apply.  It is a method recommended for use by the water utilities in Southern California who publish a monthly setback number on their website.  It is a practical form of optimization.

With that said we are looking into how to use the CIMIS data within California for full ET based cycles.

Jim