• Home
  • Feature
  • New Tools Match Wheat Varieties to Growing Environments

New Tools Match Wheat Varieties to Growing Environments

May 14, 2012

title_newtools

April, 2004

Wheat is grown in about 70 countries, in environments that extend from the Arctic Circle to near the Equator, from sea level to elevations of 4,000 meters, and under very dry and very wet conditions. Wheat researchers may not know that their local growing environment shares key limitations with environments in other parts of the world. They may not know that another scientist, half a world away, is trying to solve the same problem.

A wide-ranging project between CIMMYT and Australian organizations is helping wheat researchers obtain and share information to develop better varieties more efficiently. The project’s tools for analyzing and sharing information will enable many more researchers to work together on common problems.

CIMMYT is working with the University of Queensland (UQ) and Australia’s Commonwealth Scientific and Industrial Research Organization (CSIRO) to characterize growing environments and understand how different wheat lines grow there. (Wheat lines can be thought of as experimental, unfinished varieties.) Researchers are creating information tools—including mapping systems, wheat breeding simulation programs, and environmental simulations—that wheat researchers can use to develop more appropriate wheat varieties and production practices for a set of target environments. The project is supported by Australia’s Grains Research and Development Corporation (GRDC).

One reason that CIMMYT and the Australian organizations can benefit considerably from each other’s research tools and partnerships is that Australian wheat growing environments resemble some important wheat-producing environments in developing countries.

Testing the Ground

Part of the information that powers the project comes from the International Adaptation Trial (IAT), which consists of seed of 80 spring wheat lines of bread wheat and durum wheat. Cooperators who receive the trial plant the seed according to specific instructions, collect data from planting to harvest, and return the data to CIMMYT. CIMMYT breeders Wolfgang Pfeiffer, Richard Trethowan, Maarten van Ginkel, and Tom Payne identified cooperator sites, emphasizing sites with low rainfall and susceptibility to drought. They worked with Australian breeders to choose the CIMMYT and Australian lines that were included in the trial.

The IAT contains broadly and specifically adapted lines. Information on the performance of broadly adapted lines indicates their stability across a range of environments and in the presence of various environmental stresses, including diseases, pests, and soil problems. Individual environmental stresses are identified through specifically bred lines called probe genotypes, which have comparative responses that reflect the presence or absence of a specific trait.

In the IAT, most of the comparative pairs have highly similar genetic backgrounds, except for the trait of interest. For example, the Australian lines Gatcher and Gatcher GS50A help detect root lesion nematode. Gatcher is vulnerable to the nematode, but Gatcher GS50A is not. In the presence of the nematode, Gatcher GS50A yields better than Gatcher—more than half a ton better. In the absence of the nematode, both lines yield about the same.

Simulating the Growing Environment

The project also uses extensive sets of weather, climate, and geographical data. Along with the information from the IAT, these data are used to model how wheat lines with particular characteristics are likely to perform in key locations around the world. Running a crop simulation module that works in all types of environments is difficult, says UQ postdoctoral fellow, Ky Mathews. Researchers need good data that cover long periods. Mathews has been using daily weather data, supplied by the US National Oceanic and Atmospheric Administration, from 1973 to the present for 20,000 locations. These data are supplemented with information from cooperators. She is also using an FAO soil map to identify the most likely soil types in different regions.

From the modeling and IAT results, researchers around the world gain a more detailed understanding of target environments. They can investigate stresses at a location based on the IAT probe lines, find data on other wheat-producing locations that have similar stress responses, and evaluate weather patterns and soil information that might indicate a line’s vulnerability or exceptional resistance to a stress. This information will help breeders to make more informed choices about the lines they request from each other, the crosses they make, the genes and traits they use, and ultimately which lines they release as varieties to farmers.

It will also help them to solve shared problems. Preliminary results indicate that root lesion nematode is found at IAT sites in Ecuador, Bangladesh, India, and Mexico. Breeders can see from project maps that they experience the same challenges. “Before, we could never map the nematode sites around the world,” says Mathews. “That had never been done.”

Many Products

The project has several outputs, such as a global prediction model for flowering that defines global planting dates, a database of weather and soil data, a tool that extracts phenotypic data over the Internet from CIMMYT’s large database, and data summary tools. One tool, called QU-Cim, simulates CIMMYT’s bread wheat breeding program.

“The IAT also provides an ‘adaptation filter’ that increases the usefulness of data that CIMMYT and its partners have collected for decades in wheat breeding environments all over the world,” says CSIRO crop adaptation scientist Scott Chapman. For example, the breeders who discover a Boron problem can use CIMMYT’s historical data to identify locations where CIMMYT lines have performed well despite the presence of Boron and use these lines to develop tolerant varieties.

Mathews thinks it is important that cooperators get the project results so they can see the bigger picture. “I would like the breeders around the world to be able to have the tools to interrogate locations around the world to make better decisions about their breeding programs,” she says.

Despite the challenges, CIMMYT wheat researchers believe that the project has demonstrated tremendous potential for adding value to local and global wheat breeding research. CIMMYT is seeking funds to extend this work to more of the world’s important wheat-producing environments.

QU-Cim: Improving the local relevance of CIMMYT’s global wheat breeding programCIMMYT’s wheat breeding program has more than five decades of accumulated breeding data and has been highly successful. That makes it an excellent testing ground for QU-Cim, a tool that simulates wheat breeding processes and outcomes.

QU-Cim is a module of QU-GENE, a simulation platform developed at the University of Queensland by Mark Cooper and Dean Podlich.

QU-GENE can integrate enormous amounts of genetics-based data from widely different sources, produce realistic scenarios that help breeders compare potential outcomes without expensive field trials, and determine the best way to achieve the results they want. Only the approaches that are most likely to succeed will be used in the field.

Together with UQ programmers, CIMMYT Associate Scientist Jiankang Wang wrote the QU-Cim module and worked with CIMMYT researchers to parameterize it for CIMMYT’s breeding program.

Starting with the genetic characteristics of wheat breeding lines, QU-GENE can simulate the performance of their descendents in a given field environment over many breeding cycles. The resulting information should help breeders devise the crosses that will deliver desirable traits, even traits determined by the interaction of many genes. QU-GENE can also reduce breeding costs by reducing the number of crosses breeders make to reach a particular goal, identifying the best breeding method to use, or determining the most cost-effective, efficient time to use it.

A copy of QuCim 1.1 can be obtained by contacting either Jiankang Wang or Maarten van Ginkel.