Four Digital Information Model Of Orange Orchard

With the global climate change,extreme climate occurred frequently. The changes of environment conditions bring a worse impact to Citrus industry in recent years, it is difficult to devise, monitor and manage the orchard, built in hills. As the aging level of orchardists accelerates, informationize. Here is the result:1^ The SPSS soft software is used for modeling using stepwise regression me thod which is base of the least square method. At first we make a correlation ana lysis of whole spectral reflectance using the original spectral reflectance, finding th at there are some modeling factors both in the visible spectra and the near-infrare d spectra. Then we make two stepwise regression for the visible spectra and the n ear-infrared spectra respectively, finding that he near-infrared spectra is better for t his model. At last we get the model using several spectral pretreatments, Y=0.465-6.193×K 1528.61+9.059×K1891.31+7.161×K 1682.74-32.388xK 1837.38+22.27xK 1830.64.2,The paper studied the relationship between leaf chlorophyll content and lea f reflectance, and established models of leaves chlorophyll content based on spectr a with stepwise regression method and constructing spectral indexes method respec tively. The results showed there was good relevance between SPAD and reflection spectrum of Citrus leaves. The relative errors between predictive and observed mo dels with two methods were all less than 10%. It is clear that the two models obt ained nice forecasting results, and the model established with stepwise regression method had higher precision than the model with another method. Considering the physical meaning and logic, the model established with constructing spectral inde xes method was recommended. Y=73.491+36.718×gndvi-14.933×r.3,The paper studied the relationship between leaf water content and leaf refl cctance, and established models of leaves water content based on spectra with step wise regression method and constructing spectral indexes method respectively. The results showed there was good relevance between water content and reflection spec trum of Citrus leaves. The relative errors between predictive and observed models with two methods were all less than 10%. It made clear that models obtained bett er forecasting results, and the model established with stepwise regression method h ad higher precision than the model with another method. Considering the physical meaning and logic, the model established with constructing spectral indexes metho d was recommended. Y=-2.741 xSR-5.979×NDVI+0.62×WI2+2.122.