Note: This page was originally sourced from blogger: http://ncarrda.blogspot.com/2015/05/the-data-starts-here.html

At the Water Conservation Gardens Fair yesterday in Berthoud, Colorado (near NCAR's Boulder home), several vendors showed soil moisture monitoring systems.  I took a picture to show where data comes from.

Soil moisture monitor/data logger and one sensor.

The green-blue rectangle displayed at the top is the soil moisture sensor, which is buried in the ground.

An irregularly-spaced network of calibrated soil moisture sensors (not necessarily this brand and model) send their data to National Center for Environmental Prediction (NCEP) in College Park, Maryland. Other sensors, some in-situ like this one, others remotely sensed (usually from satellites), also send their data to NCEP.

The data is ingested into an analysis model to produce a gridded analysis field.

Some of our data users may not be aware that NCEP and NCAR are different organizations. NCEP is part of the National Oceanic and Atmospheric Administration (NOAA), which is part of the Department of Commerce (DoC). DoC headquarters are in Maryland, but they have a significant laboratory in Boulder, Colorado, home of NCAR.

Notice that we are several thousand miles and two timezones apart.

NCEP and NCAR are quite far apart in distance, but the internet brings us closer.

This is where NCEP processes the data.

NOAA NCEP Center for Weather and Climate Prediction, College Park, Md

This is where we pull and archive data from NCEP and other sources.

NCAR Mesa Lab, home of NCAR RDA

Interested in learning about how data gets calibrated?  Leave a comment with your name, organization or school, and what you'd like to read about on this blog.

Today, I took a bike ride to collect qualitative (not quantitative) data about soil moisture
and stream flow.  ;-)
The RDA's most popular dataset is FNL, ds083.2. It's popularity largely stems from its utility for initializing The Weather Research & Forecasting Model (WRF). Studies have proven that improved soil moisture estimates provide more accurate rainfall predictions.

If you use (or want to learn how to use) FNL for WRF, please read the help document, Guidance for WRF users for new NCEP GFS and FNL GRIB2 files. Notice that the Vtable (variable table) includes several soil moisture parameters at different depths.

Accurate soil moisture, soil type, vegetation and land-use characterization help produce accurate numerical weather forecasts. It takes a huge amount of data, many people, much software, and vast computational resources to produce these analyses. So, when you use the data, take a minute to think about the thousands of people who made it possible.