Saturday, 15 July 2006

Estimation for Nitrogen Status of Rice Plant by Reflectance Indices of Ground-Based Remote Sensors.

Soon-dal Hong1, Seong-Soo Kang1, Hyun-Cheol Jeong1, Sukyong Hong2, and Yi-Hyun Kim2. (1) Chungbuk National University, Department of Agricultural Chemistry, 12 Gaeshin-dong, Cheongju, Chungbuk, 361-763, South Korea, (2) National Institute of Agricultural Science and Technology, RDA, 249 Seodoon-dong, Suwon, Spain

1)Corresponding author (

The non-destructive determination of plant biomass by nitrogen stress is not possible; however crop canopy sensors that determine the normalized difference vegetation index (NDVI) have the potential to estimate living biomass and nitrogen status. Field experiment was conducted in 2005 to evaluate the effect of nitrogen (N) deficiency on rice biomass and reflectance. Nitrogen stress was imposed by implementing 4 levels of 0kg, 70kg, 100kg, and 130kg N ha-1 with 50% of basal dressing, 25% of topdressing at tiller stage, and 25% of fertilization at panicle initiation stage, respectively. Rice was planted on May 17 in clay loam soil applied with the same rate of P2O5, K2O, and SiO2, respectively for all treatments based on soil testing and was harvested on October 11, 2005. Canopy reflectance measurements were made with hand held spectral sensors including an active red and green GreenSeekerTM (Ntech Industries), a passive sensor of Crop CircleTM (Holland Scientific), an active red and amber Crop CircleTM ACS210 (Holland Scientific), and spectroradiometer (Ocean Optics models SD2000). Canopy reflectances, N content, dry weight, and panicle dry weight of rice were measured at the different growth stages from July 12 to October 11. Biomass of rice plant affected by nitrogen rate showed large differences between maximum and minimum values at the heading stage ranging from 44.6 to 64.7 g m-2 for dry weight and from 268.2 to 583.4 mg m-2 for amount of N uptake by leaves. Several reflectance indexes obtained from GreenSeekerTM, Crop CircleTM ACS210, and spectroradiometer were compared for evaluation of rice biomass and nitrogen status. The NIR/Red and NIR/Amber ratio by active Crop CircleTM ACS210 were the best tool for assessing differences in dry weight, panicle dry weight, N content and amount of N uptake by rice leaves at every growth stage evaluated. Especially, the NIR/Red and NIR/Amber ratio by active Crop CircleTM ACS210 at the panicle initiation stage were very closely correlated with dry weight, panicle dry weight, and amount of N uptake by leaves at harvesting, meaning that the panicle initiation stage is critical season for management of nitrogen fertilization for rice. Consequently these indexes were significantly explained for characteristics at the harvesting based on determination coefficient with dry weight (R2=0.82~0.83), panicle dry weight (R2=0.86), amount of N uptake (R2=0.89), LAI (R2=0.82~0.83). This result suggested that these reflectance indexes could be used for in-season N management decisions when using real-time rice sensors. Ground-based remote sensing provided a non-destructive real-time assessment of plant N status and should be a useful tool for in-season rice N management providing both spatial and temporal information. Key word; Reflectance index, ground-based remote sensor, rice, N uptake, dry weight

Back to 3.0B Emerging Topics in Soil Use and Management - Poster
Back to WCSS

Back to The 18th World Congress of Soil Science (July 9-15, 2006)