中文摘要：

为了探讨不同测量环境下碱化土壤波谱相互转换的条件和规律,对新疆天山北坡奇台绿洲碱化土壤进行定点野外、实验室波谱测量及特征分析,并结合土壤理化性质,利用多元线性回归方法进行野外-实验室测量波谱之间的双向转换。结果表明：野外测量波谱与土壤pH值呈极显著的正相关关系,可有效监测土壤碱化程度。但实验室测量波谱与pH值之间不具有相关性。pH值对实验室测量波谱向野外测量波谱的转换影响最大,其次是实验室波谱的b3（630～690nm）和b2（521～600nm）波段反射率,而b1（450～520nm）、b4（760～900nm）波段反射率与野外测量波谱的各波段反射率之间均无明显对应关系。实验室测量波谱与有机质含量呈负相关关系,且对于土壤有机质具有良好的定量反演潜力。实验室波谱向野外波谱转换的模型涉及的因子较少,稳定性高且具有良好验证精度,转换效果较野外波谱向实验室波谱转换的模型理想。

英文摘要：

To explore the conditions and rules of conversions between spectra of alkalinized soil under different measure environments,the paper investigated the laboratory-measured and field-measured spectra of alkalinized soil within the Qitai oasis at the northern slope area of Tianshan Mountain in Xinjiang.Based on the analysis of the characteristics of the laboratory-measured and field-measured spectra and physicochemical properties of alkalinized soil at the study spots,the spectral transformation between the laboratory-measured and field-measured spectra were built by multiple linear regression method.The results showed that there was a significant positive correlation between the field-measured spectra and the soil pH value,therefore soil alkalinization could be effectively monitored by using the field-measured spectra.Although the correlation between the laboratory-measured spectra and the soil pH value was insignificant,pH value has the greatest impact on the conversion from the laboratory-measured spectra to the field-measured spectra,followed by the band3（630-690 nm） and the band2（520-600 nm） of the laboratory-measured spectra.The laboratory-measured spectra of band1（450-520 nm） and band4（760-900 nm） had no significant correlations with the field-measured spectra of the band1（450-520 nm）,band2（520-600 nm）,band3（630-690 nm） and band4（760-900 nm）.The laboratory-measured spectra were significantly negatively correlated to the soil OM（organic matter）,therefore the laboratory-measured spectra had the potential for estimating the quantitative retrieval of the soil OM.The model for laboratory-to-field spectral measurement transformation was less complex with fewer variables,more stability and higher verified accuracy,so it was better than the model for filed-to-laboratory spectral transformation.