The brown planthopper, Nilaparvata lugens (Stål), a type of rice pest, immigrates from southern China to western Japan by drifting in the southwesterly environment during the Baiu/Meiyu rainy season. This study aims to investigate the preferable weather patterns for the immigration of brown planthoppers across the East China Sea. We conducted immigration runs using an advection diffusion model for brown planthoppers and projected the possibility of them landing in Kyushu in the model run onto 64 weather map patterns typically observed in June and July, identified by the self-organizing map analysis. The results showed that the immigration occurred under two specific weather map patterns: flow stagnation around western Japan and a low-level jet blowing over the East China Sea. Consistency between landing cases in model runs and the occurrence of the two weather patterns was found in the intraseasonal and interannual variability. We also estimated the change in the frequency of brown planthopper arrival in Kyushu based on the climate change dataset, the database for Policy Decision Making for Future climate change (d4PDF). It was found that the flow stagnation patterns increased in response to global change, at least in the d4PDF dataset. Finally, risk assessment for temperature change and a comparison with trap observations were discussed.

　To determine the difference in greenhouse gas emissions between two rice (Oryza sativa L.) varieties; Takanari (Tak) and Koshihikari (Kos), under organic rice cultivation with or without weeds management, a pot experiment was conducted at Tsuruoka, Yamagata, Japan. The Tak and Kos varieties were transplanted individually and interplanted as Tak+Kos, with the main treatments of weeding (w) and no-weeding (nw). Methane (CH₄), nitrous oxide (N₂O), and plant night respiration (CO₂ emission) fluxes were measured once weekly or biweekly during the rice growth period. As the results, N₂O fluxes were not detected but CH₄ fluxes varied with the rice growth stage for all treatments. Compared with the Kos(w), Tak(w) and Tak+Kos(w) significantly decreased total seasonal CH₄ emissions by 12.4% and 19.1%, respectively. This was attributed to higher rhizospheric CH₄ oxidation and the offset of potential methanogenic activities because of the larger root biomass and a higher harvest index for both Tak(w) and Tak+Kos(w) compared with Kos(w). Conversely, Kos(nw) significantly reduced cumulative CH₄ emissions by 31.7% and 48.6% compared with Tak(nw) and Tak+Kos(nw), respectively. This was ascribed to high rhizospheric CH₄ oxidation promoted by the presence of significantly higher weed biomass for Kos(nw) at 65.0% and 44.0% compared with Tak(nw) and Tak+Kos(nw), respectively. On average, the no-weeding practice significantly decreased cumulative CH₄ emissions by 50.5% compared with the weeding practice, with a significantly high (P<0.001) interaction between weed management and variety differences observed. The CH₄ emissions and plant biomass under variety interplanting (Tak+Kos) were not significantly different to Tak individual planting, but were significantly different to Kos individual planting, which implied variety Tak had strong competition to variety Kos. More studies on appropriate variety choice and pairs, and the effect of aquatic weeds presence in no-weeding organic paddy fields are necessary to estimate CH₄ emission and attenuation.

　A custom-made fluorescence measurement system was constructed for non-destructive and quantitative estimation of green fluorescent protein (GFP) transiently expressing in a Nicotiana benthamiana leaf with a viral vector. The system comprised a digital camera, a blue LED lamp, and a long wavelength-pass filter to acquire the brightness intensity of the green channel (G value) of a leaf surface image as a measure of fluorescence intensity. The genes encoding GFP and tobamoviral vector were transferred to N. benthamiana leaves 42 days post seeding by syringe infiltration with Agrobacterium for transient expression. The destructively measured GFP content in leaves as the ground-truth data started to increase at 3-4 days post infiltration (DPI), once showed a plateau at 6-8 DPI, and then increased again until 11 DPI. Regression analysis demonstrated that the r² of the linear regression between the ground-truth GFP content and the G value was 0.86-0.87 when the G value at a given DPI was calibrated with the initial leaf color at 0 DPI for the same leaf. These results indicate that the constructed system and method could potentially be used to quantitatively and non-destructively estimate the amount of GFP transiently expressing in intact N. benthamiana leaves.