Estimation of Nitrogen Content Based on the Hyperspectral Vegetation Indexes of Interannual and Multi-Temporal in Cotton

Crop nitrogen is an efficient index for estimating crop yield. Using hyperspectral information to monitor in cotton real time can help guide cultivation. In this study, we used drip-irrigation Xinjiang as the research object and employed various treatments explore correlation between vegetation indexes leaf concentration (LNC) canopy density (CND) of different growth periods interannual. We 30 published obtained through spectral monitoring 2019 2020 screen highly correlated with indexes. Based on same group indexes, interannual multi-temporal estimation models were established using three modeling methods: simple multiple linear regression (MLR), partial least-squares (PLSR), support vector (SVR). The results showed following: (1) correlations LNC CND individual lower than those entire period. LNC, CND, area (LAI), aboveground biomass (AGB), significantly years varieties. relatively stable TCARI, PRI, CCRI, SRI-2, absolute values 0.251~0.387, 0.239~0.422, 0.245~0.387, 0.357~0.533. addition, CIred-edge REIlinear indicators (CND, AGB, LAI) was more stable. (2) by MLR, PLSR, SVR, R2 value from SVR method higher model based period data CND. (3) selected estimate stages, accuracy that concentration. most MLR at flowering boll stages full-boll stage = 0.532~0.665. This study explored application potential drip-irrigated cotton, provide a theoretical basis nutrients structure.