Improving Fertilizer Guidelines for a Changing Rice Climate, 2017


Project Leader

Bruce Linquist, UCCE specialist, Dept. of Plant Sciences, UC Davis

The goal of this project is to develop fertilizer management guidelines that are economically viable and environmentally sound. Research objectives in 2017 were:

  • Determine the potassium status of rice soils.
  • Develop management practices for growing rice under alternating flooded/dry conditions.
  • Quantify rice yield variability in the Sacramento Valley.
  • Improve fertilizer nitrogen use efficiency.

Potassium status

The main focus of research in 2017 was to complete this study by determining mineralogy of the soil and how that relates to potassium fixation.

In 2012 and 2013, 55 rice fields were identified in the Sacramento Valley for potassium status research. Samples were taken from the soil, water, and rice flag leaves in all fields and analyzed for potassium. In addition, growers were asked about yields, straw management, and winter flooding practices.

Soil potassium values ranged from 35 ppm to 350 ppm. There was no relationship between soil potassium values and the amount of potassium that had been added and removed. Soil potassium values were lowest in the southeast part of the valley, followed by the northeast and northwest. Highest values were in the southwest. All fields below the threshold of 60 ppm were on the east side.

When soil potassium values were below 60 ppm, half the flag leaves sampled had values below the critical level of 1.2%. Where soil potassium ranged between 60 ppm and 120 ppm, eight flag leaf samples had potassium

levels below the critical range. Thus, potassium fertilizer should be considered when soil potassium levels are below 120 ppm.

There was a significant difference in the concentration of potassium in irrigation water. The Sacramento River had the highest potassium values (1.18 ppm), while the Feather River averaged 0.79 ppm. Well water had the highest overall potassium concentration (2.3 ppm), but it was also highly variable. Recycled irrigation water averaged 1.4 ppm and also was variable.

Some soils low in available potassium were also potassium fixers. That is, these soils could bind potassium in forms that are not available to plants. They are located mostly east of the Sacramento River and require a different potassium fertilizer regime. Differences in potassium fixation among soils is likely caused by differences in soil mineralogy. Additional research will lead to refined potassium fertilizer recommendations.

Alternating wet/dry rice

Current water management practices keep most California rice fields continuously flooded through the majority of the growing season. This strategy helps provide high yields, good weed control, and efficient nitrogen use. Nonetheless, there is interest in exploring alternative production practices such as alternating flooding with periods of dry soils. This practice goes by the acronym AWD.

An ongoing study at the Rice Experiment Station has been evaluating this practice. In all AWD treatments, grain yields were the same as in the conventional water-seeded treatment. Similarly, the optimum nitrogen rate required to achieve maximum yields was similar among treatments. This research also showed that AWD reduces greenhouse gas emissions by 57% or more and grain arsenic by 50% (as well as methylmercury). “Safe” AWD, which allows soils to dry for two days before reflooding, reduced methane emissions by 40%.

While these results are encouraging, it is not clear how easy this method would be to implement at the field scale. This research suggests a field can be drained without a yield penalty for up to 12 days, depending on location and soil type.

Rice yield variability

California rice yields are among the highest in the world. However, over the past 15 to 20 years, yields have stagnated. The goal of this area of research is to identify ways to further increase yields through improved management.

Research has focused on finalizing a yield gap assessment of U.S. rice systems and determining the primary temperature stresses that impact rice yields in California. Data have been compiled into a central database and yield information linked to site-specific daily climate and soils data. A procedure was developed to calculate climate variables, such as average temperature during flowering, that is specific to each data point in the statewide variety trials.

A model (ORYZA) to estimate yield potential produced the following results:

  • The current maximum yield potential in California rice systems with modern medium grains is between 11,200 pounds/acre and 12,500 pounds/acre, while the Southern U.S. generally has lower yield potential. Actual statewide yields in California average between 7,700 to 8,600 pounds/acre. The yield gap—where there is room for potential improvement—is in the range of 1,000 pounds/acre to 1,500 pounds/acre in California.
  • The impact of cool stress during booting had the largest impact on grain yield in the model—up to 2,500 pounds/acre. Cooling or heat stress during flowering were found to have the next largest impact on grain yield. Warmer seasonal temperature minimums and maximums were found to have a much smaller impact on yield—less than 400 pounds/acre.
  • Sensitivity to cooling and heating stress during booting varied by grain type, with medium and short grains being more resistant to cooling stress and long grains being more sensitive. Sensitivity to cool stress during flowering was similar among grain types. Long grains were least sensitive to heat stress during flowering compared to short and medium grains.

Midseason nitrogen status

The objective of this study was to examine the potential of using remotely sensed data to determine the need for a midseason top-dressed nitrogen application.

A handheld sensor was evaluated in multiple farmer fields in 2015, 2016, and 2017 to improve the accuracy of estimates. Additionally, a drone was incorporated into the study. Nitrogen response trials were established in three locations with seven nitrogen rates ranging from zero to 250 pounds/acre. Data from the three years of the study were compiled. The handheld sensor was not a good predictor of biomass at panicle initiation and of nitrogen concentration in the plant.

However, sensor readings did correlate well with aboveground nitrogen content. Preliminary analysis shows that if nitrogen uptake at panicle initiation is 120 pounds/acre to 130 pounds/acre, the Green Seeker NDVI (normalized difference vegetation index) value would be 0.75. An NDVI value lower than this indicates the need for a top-dressed nitrogen fertilizer. Data from this research is still being analyzed. However, given these initial promising results, along with the increased availability and use of remotely sensed data, further research in this area is warranted in order to make these tools more promising for growers.