基于多时相棉花长势遥感的棉田质量诊断

Diagnosing Cotton Field Quality with Multi-Temporal Remote Sensing Data of Cotton Growth

【目的】研究多时相遥感信息对棉田质量进行诊断的技术和方法,为棉花生产提供具有针对性的管理方案,促进棉田均衡增产、增效。【方法】研究分析多时相遥感数据对地物信息的动态分析与评判能力,以及棉花长势指标动态变化与棉田质量的关系,对棉花生长盛期多时相的LANDSAT-5多时相遥感数据进行融合,将棉田质量状况划分为健康棉田、有障碍棉田和疑似有障碍棉田三类。【结果】数据分析结果表明,棉花生长盛期(花铃期)单时相的LANDSAT-5反射率数据可以作为棉田生长状况的判断指标,划分健康生长与生长障碍的阈值为0.820;利用多时相遥感数据的棉田质量划分方法,可以将棉田质量分为健康、有障碍和疑似有障碍三类。依据此方法对新疆建设兵团148团约11705.3ha的417块棉田进行分类,得出健康、有障碍和疑似有障碍三类棉田所占比例分别为36.4%、34.1%和29.5%;经过8块条田(426.5ha)的地面同步调查证实了这种方法的准确性,造成该区棉田质量障碍的主要因素为耕地盐渍化、不平整、土壤质地不匀。【结论】研究表明棉田多时相遥感数据进行棉田质量诊断是可行的。利用这种方法,结合不同质量棉田形成机理,能够得到棉田质量状况及影响因素精细分布的信息,为进一步进行的棉田土壤改良与生产管理采取针对性的措施提供数据支撑。

[Objective] The cotton field quality was diagnosed with the remote sensing technology, and the results would provide the technology support to take the active measurements for cotton industry, and would promote to increase the yield and efficiency. [Method] In terms of the determination ability of multi-temporal remote sensing data for dynamic information, and the relationship between cotton growth and cotton field quality, the multi-temporal remote sensing of the flower-boll stages data was fused in the years form A.D. 2005 to A.D.2006. The cotton field quality conditions were divided into the three styles of the healthy cotton field, handicapped cotton field and suspected cotton field with handicap. [Result] The results showed that the 0.82 of LANDSAT-5 TM4 reflective was reasonable to divide the healthy and handicapped cotton fields with the single-time remote sensing images in the flower-boll stages of cotton, and then 417 cotton fields about 1 1705.3 ha was classified using the method with the multi-temporal data, the results were that the three-style proposition of cotton field quality was 36.4%, 34.1% and 29.5% respectively; the validity of classification was proved by the synchronization investigation based on the eight cotton field （426.5 ha）, and the testing results of soil character and total salt indicate that the main factors of handicapped cotton field were salting, disunity of character, difference of level. [Conclusion ] It was believed to diagnose the cotton field with multi-temporal remote sensing data of cotton growth and map the information of cotton field quality on the base of the its mechanism, the precise information of cotton field quality would provide the data support to improve the cotton soil conditions.