Soil probe network makes big data a big tool

GroundCover Live and online, stay up to date with daily grains industry news online, click here to read more

A national network of probes in the US gives growers real-time soil information

Image of remote monitoring system

John Deere’s Field Connect system in the US continuously logs soil moisture data from probes and transmits the data to a website.

PHOTO: Greg Baker

Hydrologists in the US are creating a national network of soil-moisture information in a multi-agency collaboration effort – similar to GRDC’s Water Use Efficiency Initiative.

Image of Craig Neild

Data collection and customisation specialist Craig Neild from Farmlink Rural installing soil moisture probes at a Liebe Group site in WA.

PHOTO: GTBA

The US Department of Agriculture’s (USDA) Dr Michael Strobel gave an insight into the intricacies of data integration when he spoke at the Soil, Big Data and the Future of Agriculture Conference in Canberra in June.

The conference, an initiative of the United States Studies Centre at the University of Sydney, looked at strategies to harvest and use vast amounts of information – or ‘big data’ – collected by digital technologies.

It is a challenge being tackled by scientists from NASA, the USDA, Texas A&M University and other agencies in the pilot phase of the US National Soil Moisture Network, which began in January 2015.

Dr Strobel, who is director of the USDA’s Natural Resources Conservation Service’s National Water and Climate Center in Portland, Oregon, said the catalyst was the need for reliable, consistent and integrated soil information to forecast and manage droughts and floods on a national scale.

“The US experiences similar climatic challenges to Australia, which has increased awareness of the implications of variable soil conditions,” he said. “By collating soil-moisture data and presenting it in real time, we can assess drought conditions and flood potential, estimate crop yields, and forecast water supply and the impacts of climate change.”

As in Australia, it is, however, easier said than done when a large number of state and federal soil data collection services need to be brought together.

Dr Strobel oversees two of these national data networks. The Soil Climate Analysis Network’s 221 stations measure and transmit data on soil moisture, soil temperature and climate. The Snow Survey and Water Supply Forecasting Program operates more than 880 automated telemetry stations and manages 1100 manual snow courses in the western US to produce water-supply forecasts.

Image of Michael Strobel

Soil, Big Data and the Future of Agriculture Conference speaker Dr Michael Strobel.

PHOTO: Mark Graham

There are also remote-sensing tools such as the National Oceanic and Atmospheric Administration’s (NOAA) microwave and thermal infrared observations, NASA’s Soil Moisture Active Passive satellite mission (which is measuring and mapping the Earth’s soil moisture) and the University of Arizona’s Cosmic-Ray Soil Moisture Observing System.

Throw in major land surface models and multi-model approaches, such as the North American Land Data Assimilation Systems (between NASA and NOAA), and it creates what Dr Strobel describes as a “data-rich, data-challenged” scenario.

The data collected is highly variable. It is difficult to calibrate simulation models to in situ measurements as the variety of sensor types and depth, spatial distribution, time frames and format of data storage produce different sets of data.

For example, some sensors collect data in dryland areas, others from irrigated soils. Sensors measure soil moisture at different depths (5, 10, 50 and 100 centimetres) and others incorporate surface conditions such as precipitation, humidity and temperature.

Remote-sensing tools also have different resolutions, making integration difficult.

In addition, some station metadata is not easily accessible online, sites can be added or removed over time and standardised quality control would be required to provide ‘clean’ data.

Dr Strobel told the conference a fully integrated soil-moisture network would contribute to water use efficiency and economic savings by giving growers information for decisions such as variety selection, time of planting and inputs.

Including snow data into the soil-moisture network would benefit a range of stakeholders, including hydropower plants and recreational organisations.

“The 12 western states in the US rely on snow melt for most of their annual stream-flow,” Dr Strobel explained. “Snow sensors provide information around the timing and volume of water, which ties into the soil moisture network by forecasting infiltration and run-off.”

Dr Strobel said the pilot study, which ended in August, would report on the value of soil-moisture data to a broad range of end users and identify best practices to calibrate data on a larger scale.

He hoped the next step would be to develop a map-based visualisation decision-making tool for growers, hydrologists and government agencies.

Dr Strobel said that although there had been many challenges in the pilot phase, he was encouraged by the collaboration and partnerships formed by the agencies involved. He felt the message for the Australian industry, if it goes down the same path, is to recognise the value in communication and collaboration: “Sharing expertise, resources and a common purpose produces a better product.”

More information:

Dr Michael Strobel,

michael.strobel@por.usda.gov

Next:

Pulse confidence continues

Previous:

On the frontline of IWM

GRDC Project Code GOG00009

Region National