面向水源地保护的地块尺度桉树遥感识别方法研究
|
摘要
速生桉人工林具有生长速度快、轮伐周期短的特点,大量、单一的速生桉种植给水源地的水土保持及生物多样性带来潜在威胁。为加强饮用水水源保护区环境管理,开展了面向水源地保护的地块尺度桉树遥感识别方法研究。以广西南宁市重要饮用水保护区——大王滩水库为对象,基于高分一号16 m分辨率影像提取库区水域信息和边界,进而向外缓冲500 m获得研究区范围;然后以谷歌高分辨率影像为数据源,利用人机交互方法获取研究区内的精细地块信息;最后结合多时相中分遥感影像计算地块的多维光谱特征,并利用基于地块的面向对象分类方法提取桉树、其他林地、耕地、建设用地等土地覆盖信息。结果表明,基于地块尺度的桉树识别方法能够获得较好的提取效果,总体分类精度为90.24%,Kappa系数为0.88,其中桉树识别的用户精度及生产者精度分别达到94.87%和97.37%,能够满足应用需求。研究发现桉树种植面积高达4 960.03 hm~2,占研究区面积的35.02%,面积比例远高于其他树种,可能对水源区生态环境带来不利影响,建议有关部门加强对水源区桉树种植及砍伐的管控,确保饮用水保护区的生态环境质量。
The fast-growing Eucalyptus had high growth rate and short cutting period, the large-scale eucalyptus plantation threatened the biodiversity, soil and water conservation of water reservation area.In order to facilitate the environmental management of water reservation areas, this paper analyzed the remote sensing identification of spot scale Eucalyptus for drinking water source protection. The Dawangtan Reservoir,one of the most important drinking water conservation areas in Nanning was chosen as the study object in this paper. The water area of reservoir was extracted by 16 m GF-1 image, a 500 m-width buffered area surrounding the water-body was made and served as study area. Based on the Google' s high-resolution imagery, the land information were produced by human-computer interaction. The multi-dimensional spectral features of land was calculated by multi-temporal mid-resolution images. The eucalyptus, woodland,cultivated land and construction land were identified by the object-oriented classification method. The results showed that the remote sensing identification of spot scale Eucalyptus had better extraction effect, the overall accuracy was 90.24%, the Kappa coefficient was 0.88. The user and producer accuracies of Eucalyptus identification reached 94.87% and 97.37%, this could adequately meet the application needs. The planting area of eucalyptus reached 4 960.03 hm~2, accounted for 35.02% of the study area,which was much larger than other trees. Considering the worse effect on the environment of water reservation area,it was suggested that management departments should strictly regulate the planting and deforestation of Eucalyptus in order to ensure the environmental quality of drinking water protection area.
The fast-growing Eucalyptus had high growth rate and short cutting period, the large-scale eucalyptus plantation threatened the biodiversity, soil and water conservation of water reservation area.In order to facilitate the environmental management of water reservation areas, this paper analyzed the remote sensing identification of spot scale Eucalyptus for drinking water source protection. The Dawangtan Reservoir,one of the most important drinking water conservation areas in Nanning was chosen as the study object in this paper. The water area of reservoir was extracted by 16 m GF-1 image, a 500 m-width buffered area surrounding the water-body was made and served as study area. Based on the Google' s high-resolution imagery, the land information were produced by human-computer interaction. The multi-dimensional spectral features of land was calculated by multi-temporal mid-resolution images. The eucalyptus, woodland,cultivated land and construction land were identified by the object-oriented classification method. The results showed that the remote sensing identification of spot scale Eucalyptus had better extraction effect, the overall accuracy was 90.24%, the Kappa coefficient was 0.88. The user and producer accuracies of Eucalyptus identification reached 94.87% and 97.37%, this could adequately meet the application needs. The planting area of eucalyptus reached 4 960.03 hm~2, accounted for 35.02% of the study area,which was much larger than other trees. Considering the worse effect on the environment of water reservation area,it was suggested that management departments should strictly regulate the planting and deforestation of Eucalyptus in order to ensure the environmental quality of drinking water protection area.
引文
[1]付晓,王雪军,韩爱惠,等.纸浆源地桉树资源遥感动态监测与分析[J].北京林业大学学报,2008,30(S1):89-93.
[2]黎洁,温亚,何蓉,等.桉树种植的现状、生态问题与可持续发展对策研究[J].绿色科技,2017(2):94-97.
[3]陈少雄,陈小菲.我国桉树经营的技术问题与思考[J].桉树科技,2013,30(3):52-59.
[4]李志辉,杨民胜,陈少雄,等.桉树引种栽培区区划研究[J].中南林学院学报,2000,20(3):1-10.
[5]Engler R,Waser L T,Zimmermann N E,et al.Combining ensemble modeling and remote sensing for mapping individual tree species at high spatial resolution[J].Forest Ecology and Management,2013,310(1):64-73.
[6]Christopher E,Churches Peter J.Wampler,Wanxiao Sun,et al.Evaluation of forest cover estimates for Haiti using supervised classification of Landsat data[J].International Journal of Applied Earth Observation and Geoinformation,2014,30:203-216.
[7]伍静.基于TM的速生桉遥感分类技术研究[D].南宁:广西大学,2014.
[8]蔡耀君.基于面向对象技术的Rapid Eye影像桉树林信息提取研究[D].南宁:广西师范学院,2015.
[9]梁文海,刘吉凯,张伟,等.基于面向对象方法的GF-2影像桉树林信息提取[J].浙江农林大学学报,2017,34(4):721-729.
[10]谢国雪,杨如军,李双青,等.基于空间三维视角的无人机技术监测农林资源[J].亚热带资源与环境学报,2015,10(2):84-90.
[11]谢国雪.无人机倾斜摄影系统的三维可视化应用研究[D].南宁:广西师范学院,2016.
[12]覃泽林,谢国雪,李宇翔,等.多时相高分一号影像在丘陵地区大宗农作物提取中的应用[J].南方农业学报,2017,48(1):181-188.
[13]牟智慧,杨广斌.基于TM影像面向对象的桉树信息提取[J].林业资源管理,2014(2):119-125.
[14]王学成,杨飞.基于面向对象分类方法的Rapideye桉树提取[J].桉树科技,2014,31(3):10-16.
[2]黎洁,温亚,何蓉,等.桉树种植的现状、生态问题与可持续发展对策研究[J].绿色科技,2017(2):94-97.
[3]陈少雄,陈小菲.我国桉树经营的技术问题与思考[J].桉树科技,2013,30(3):52-59.
[4]李志辉,杨民胜,陈少雄,等.桉树引种栽培区区划研究[J].中南林学院学报,2000,20(3):1-10.
[5]Engler R,Waser L T,Zimmermann N E,et al.Combining ensemble modeling and remote sensing for mapping individual tree species at high spatial resolution[J].Forest Ecology and Management,2013,310(1):64-73.
[6]Christopher E,Churches Peter J.Wampler,Wanxiao Sun,et al.Evaluation of forest cover estimates for Haiti using supervised classification of Landsat data[J].International Journal of Applied Earth Observation and Geoinformation,2014,30:203-216.
[7]伍静.基于TM的速生桉遥感分类技术研究[D].南宁:广西大学,2014.
[8]蔡耀君.基于面向对象技术的Rapid Eye影像桉树林信息提取研究[D].南宁:广西师范学院,2015.
[9]梁文海,刘吉凯,张伟,等.基于面向对象方法的GF-2影像桉树林信息提取[J].浙江农林大学学报,2017,34(4):721-729.
[10]谢国雪,杨如军,李双青,等.基于空间三维视角的无人机技术监测农林资源[J].亚热带资源与环境学报,2015,10(2):84-90.
[11]谢国雪.无人机倾斜摄影系统的三维可视化应用研究[D].南宁:广西师范学院,2016.
[12]覃泽林,谢国雪,李宇翔,等.多时相高分一号影像在丘陵地区大宗农作物提取中的应用[J].南方农业学报,2017,48(1):181-188.
[13]牟智慧,杨广斌.基于TM影像面向对象的桉树信息提取[J].林业资源管理,2014(2):119-125.
[14]王学成,杨飞.基于面向对象分类方法的Rapideye桉树提取[J].桉树科技,2014,31(3):10-16.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |