Near infrared reflectance spectroscopy (NIRS) is routinely used for the determination of nutrient components of forages. However, little is known about the impact of sample preparation and wavelength on the accuracy of the calibration to predict minerals. This study was conducted to assess the effect of sample preparation and wavelength of near infrared spectrum for the improvement of calibration and prediction accuracy of Calcium (Ca) and Phosphorus (P) in imported hay using NIRS. The samples were scanned in reflectance in a monochromator instrument (680–2,500 nm). Calibration models (n = 126) were developed using partial least squares regression (PLS) based on cross-validation. The optimum calibrations were selected based on the highest coefficients of determination in cross validation (R2) and the lowest standard error of cross-validation (SECV). The highest R2 and the lowest SECV were obtained using oven-dry grinded sample preparation and 1,100-2,500 nm wavelength. The calibration (R2) and SECV were 0.99 (SECV: 468.6) for Ca and 0.91 (SECV: 224.7) for P in mg/kg DM on a dry weight, respectively. Results of this experiment showed the possibility of NIRS method to predict mineral (Ca and P) concentration of imported hay in Korea for routine analysis method to evaluate the feed value.

ABSTRACT
Ⅰ. 서 론
Ⅱ. 재료 및 방법
    1. 시료 및 NIR 스펙트라 수집
    2. 수입건초의 Ca 및 P 함량 분석
    3. NIR 검량식 작성
Ⅲ. 결과 및 고찰
    1. NIR 스펙트라의 특성
    2. 수입건초 시료집단의 Ca 및 P 함량 범위
    3. 검량식 작성 및 검증
Ⅳ. 요 약
Ⅵ. REFERENCES

• 이배훈(국립축산과학원) | Bae Hun Lee (National Institute of Animal Science, RDA)
• 김지혜(국립축산과학원) | Ji Hye Kim (National Institute of Animal Science, RDA)
• 오미래(국립축산과학원) | Mirae Oh (National Institute of Animal Science, RDA)
• 이기원(국립축산과학원) | Ki Won Lee (National Institute of Animal Science, RDA)
• 박형수(국립축산과학원) | Hyung Soo Park (National Institute of Animal Science, RDA) Corresponding author