What can the Climate Outcome Likelihood Tool tell me about future
This tool presents the distribution of accumulated precipitation values
during the specified recovery period for all like periods observed in
a station's record. It can tell you how likely an outcome (such as
recovering a deficit and reaching normal or exceeding a certain
precipitation threshold) is based on historic data. This tool can be
used to bound decision making based on historical precipitation
accumulations. For example, say 5 inches of precipitation are needed
to recover a deficit/reach normal at a particular station between
March 1 and May 31. If the station's record is 100 years long and
only 6 March 1-May 31 periods exceeded 5 inches, there would be a 6%
chance of recovering the deficit and reaching normal in the March 1
to May 31 recovery period. Information on interpreting an example
graph and table is shown below:
What does the Climate Outcome Likelihood Tool NOT tell me
about future precipitation outcomes?
This tool does not incorporate any weather forecast models and
therefore it is not a forecast tool. This tool does not incorporate
any climate model data and therefore does not account for
non-stationarity of the climate as we move forward in time. Extreme
precipitation values beyond the bounds of the observed historic data
presented here are certainly possible and should be considered in any
planning process. This tool helps to put the distribution of
precipitation at a station in context and display the current bounds
of the data but should not be used alone as a forecasting tool.
Why was this tool developed?
The Climate Outcome Likelihood Tool was developed to help answer the
question, "How might this year turn out based on historic
precipitation?" The question rose out of the ongoing California
drought beginning in 2011 as people wondered at the chances of getting
out of drought and/or recovering a precipitation deficit. The tool was
developed with water managers in California in mind but is open to
everyone and any GHCN station ID in the US can be analyzed.
Does recovering a precipitation deficit end a drought?
Not always. Drought is a multi-faceted issue and is often strongly
tied to its impacts. In some cases, drought can be ameliorated
without recovering a long-term precipitation deficit. Conversely, a
precipitation deficit may be recovered but the precipitation was
received all in one large dose and ecosystems and resources were not
able to respond in a way that suggests drought recovery. NCDC operates a tool that allows users to look
at probability of ending drought conditions based on drought
indicators beyond precipitation.
I don't like the decile, normal, amount needed lines on the
images. Can I get rid of them?
No. We were hoping to design the graphics so we could turn these lines
on and off but were limited by the graphing software chosen
Why can I only see stations in California on the map?
This tool was developed for the California Department of Water
Resources. It can be used for any GHCN station in the US by simply
entering that station's ID in the station ID area. You can find
station IDs on WRCC's
SCENIC station finder or
WRCC COOP Station
Why are stations with a record 1920 or earlier to present
The major droughts in California weather station recorded history took
place in the late 20s/early 30s, 50s, mid-70s, and late 80s/early 90s.
Since this tool was born out of California drought issues, we wanted
to make sure distributions presented included several major droughts.
Additionally, stations with a record 1920-present generally have 95+
years (though may have some missing years in record) of data. This
allows for a more meaningful distribution to be generated than if the
station had only 20 or 30 years of data. The analysis can be
performed for stations of any record length by entering station ID.
You can find station IDs on WRCC's
station finder or
WRCC COOP Station
Why is the middle date set to "today"? What happens if I change
The middle date by default is set to "today" on the assumption that
people are interested in the likelihood of outcomes from today
looking forward, the period for which data has not yet been observed.
If you would like to look at a future period that begins beyond today,
say for example it is April 1, 2015, and you are only interested in
possible precipitation for September 1-December 31, 2015, that works
as well. If you choose a middle date prior to today, unless
you change the missing days allowed in each recovery period to a high
enough value to accommodate the number of missing days between the
middle and end date, the value for the most recent period will not be
included as a recovery period due to too many missing values.
Why am I limited to +/- 5 years in this tool?
For the results of a query to be processed in an amount of time that
users would likely be willing to wait, we found it best to restrict
the time period to 5 years into the past and out into the future.
This allows for inclusion of the precipitation deficit accrued in the
currently 4 year (as of 2015) drought in California.