The rice water weevil, Lissorhoptrus oryzophilus Kuschel, is the most destructive insect pest of rice in the United States. Adult weevils move into rice fields when the fields are flooded, feed on the upper leaf surfaces producing longitudinal scars, and attain peak population density 4~ 18 (mean = 7.8 days) after permanent flooding. Females deposit eggs in leaf sheaths below the water surface. Maximum oviposition occurs 1~ 2 weeks after flooding, and larvae hatch in 4~ 9 days, depending on temperatures. Newly hatched larvae mine leaf sheaths for a short period then move to the soil to feed on roots.

Larval populations reach peak levels » 2~ 3 weeks after the peak of adult feeding, and normally » 3~ 4 weeks after permanent flooding. Larval feeding can be economically significant and result in stunted plants with few tillers and low grain yield. Four larval instars are produced, which collectively require » 21 days to complete development. Adults fly to hibernation sites as early as July, where they enter diapause and overwinter.

The objective of this study was to quantify the impact of rice water weevil injury on rice plant growth, development, and grain yield using a physiologically-based rice population simulation model (RICEPSM). The approach was to couple field larval density estimates to RICEPSM, and to use a statistically rigorous procedure to parameterize, verify, and validate the model using data obtained from the literature.

The seasonal (1989-1992) larval age-structured population densities were estimated from published field data from experiments conducted at the Louisiana Agricultural Experiment Station, LA (Quisenberry et al. 1992; Thompson et al. 1994). These estimates, the related agronomic variables, and weather data were used as input to RICEPSM to simulate the response of rice to rice water weevil injury. Two mechanisms were used to represent the effect of larval feeding.

These were a reduction in root mass through pruning and a reduction in the ability of the remaining but partially injured roots to take up nitrogen. A statistically based parameterization verification-validation procedure was used to quantify the robustness of the model at simulating rice straw mass, grain yield, and total above ground mass. The model accurately simulated each of these variables for the range of rice water weevil seasonal age-structured population densities, and explained 95% of the yield variability in the observed data.

Analyses of the simulated seasonal patterns of tiller density, and mass of root, stem, leaf, and grain using 1989 and 1992 rice water weevil data for both the untreated control and the single carbofuran application treatment, and for the rice model in the absence of larval injury indicated that the stage of plant growth, during which larval injury occurred, affects crop sensitivity to root injury. In 1989, the rice water weevil population began to increase » 480 degree-days (DD) from planting, with the carbofuran treatment receiving an application at 580 DD. At that time, the rice plants had 6 leaves, tillers were beginning to emerge, and plant mass was increasing rapidly. The 1989 larval population densities in both the untreated and single carbofuran application treatments appeared to have had no effect on tiller production and leaf mass, and only a limited effect on stem mass and grain yield, even though the injury in both treatments appeared to have greatly reduced root mass.

The limited response of the rice crop to the injury appears largely the result of the crop being sink limited at that stage of crop growth. In contrast, the weevil population began to increase at 350 DD from planting in 1992, with the carbofuran treatment receiving an application at 400 DD. At this stage, the rice plants had only 3 to 4 leaves and only a small amount of root mass had developed. The weevil population density in the untreated control would account for a high proportion of the root mass being eaten, which would have had a large affect on rice growth and development.

RICEPSM when linked to rice water weevil density data provides a robust means with which to address the development of a dynamic economic threshold.

Document Author:

Revised: