Date of Graduation

12-2022

Document Type

Thesis

Degree Name

Master of Science in Crop, Soil & Environmental Sciences (MS)

Degree Level

Graduate

Department

Crop, Soil & Environmental Sciences

Advisor/Mentor

Roberts, Trenton L.

Committee Member

Daniels, Michael B.

Second Committee Member

Hardke, Jarrod T.

Third Committee Member

Richardson, Michael D.

Keywords

nitrogen; rice; soil

Abstract

Proper nitrogen (N) management in rice production can be difficult to achieve without the aid of tools such as soil and tissue tests. There are no tests currently available to rice producers that determine whether N has been managed properly throughout the season or that detect instances of over fertilization. Rice stalk samples were collected from N response trials within 3 days of harvest from the primary Arkansas rice growing region in 2016-2018. Stalk samples were analyzed for N concentration through a KCl extraction and spectrofluorometric quantification. Samples analyzed with leaf material intact contained statistically different N concentrations than samples without leaf material (P < 0.0001). Although a large amount of variability was observed in N concentrations measured in the form of NH4+ and amino acids, N measured in the form of NO3- were more consistent in response to N rates. Samples collected lower on the rice stalk contained higher concentrations of N under excessive N fertilization conditions (P < 0.0001) indicating these samples are more tightly linked to N fertilization rates. These results suggest that rice stalk samples collected at the end of the growing season from the lower portion of the plant and analyzed for NO3- could be used to effectively determine if excess N was available to the rice plant. To further examine these results, additional samples were taken, analyzed in the same manner, and used to establish the relationship between stalk-N concentration and grain yield. Two models were used to describe this relationship: the linear plateau model and the quadratic plateau model. The linear plateau model was found to represent the data better than the quadratic plateau model (F-test: P < 0.0001, mean absolute error: 8.08 vs 8.27, RMSE: 0.128 vs 0.131, AIC: 3409 vs 3453). The join point of the linear response region and plateau region represents the critical tissue concentration at which no increase in yield is observed from addition N input. To promote fewer false negatives and ensure yield is not limited, the upper limit of the 95% confidence interval (2.1 mg NO3--N kg-1) was selected as a practical agronomic threshold for the post-season tissue test. Rice producers can use the results of this tissue test to improve N management in future growing seasons or this test can also be used to either confirm or eliminate N as the cause of undesirable yields for the current growing season.

Download

Share

COinS