在地学信息图谱思想和数字带谱体系的基础上,阐明了垂直带谱数字识别的概念和方法.通过AML语言组合ArcGIS功能模块,实现了昆仑山北侧垂直带谱的数字识别,通过评价认为其成果能够适用于山系尺度的带谱研究;探讨了Matlab软件编程实现山体360°带谱数字识别的算法;进一步分析了垂直带研究中带谱界限自动识别的前景和挑战.在对两种程序设计方法进行对比和分析的基础上,确立了山体单侧、山体360°两种垂直带谱模式的构建技术.研究表明,带谱数字识别技术能够为山地垂直带谱的构建提供可行的支持,促进地学信息图谱的进一步发展.

Geo-infoTupumethodology is a new explored direction forgeography, and it snotonly away forpresentation\nbut also amethod ofanalysis in geo-information science. Mountain altitudinalbelts (MABs) are distributed atdifferentel-\nevations in highmountains, mainly owing to varying clmi atic conditions as altitude increases. TheMABs is combined in\nvarious vertical“spectra”, and is the representativeGeo-infoTupu because of its characteristics, but its data sources and\npresentation patterns are lmi ited by the classical researchmethods. It is difficult to dealwith altitudinalbelts and to illus-\ntrating their patterns atdifferentscaleswith traditional techniques. In this paper, we discuss the digitalization and visual-\nization of theMABswith a view of fulfilling the identification of theMABs in theway of computerization.\nBased on the idea and system ofGeo-infoTupu and digitalmountain altitudinalbelts, thispaperdiscusses the digital\nidentificationmethod of theMABs and constructs feasible technicalprogrammes. The data sources of the digital identifica-\ntion of theMABs are vegetation maps, and the products of our arithmeticswould be distribution patterns of altitudinal\nbelts. The paper provides twomodels of digital identification for theMABs at different geographical scales, such as the\nsingle flank pattern and the single peak pattern. We have fulfilled the digital identification of theMABs in the northern\nflank of theKunlunMountains with the arc macro language (AML), and the result responds to the structure-thinning\nmodelwhich hasbeen validated by some scientific reports. The arithmetic ofsingle flank is applicable in the long and nar-\nrowmountain system, which could provides the general information and distribution patterns of the altitudinal belts. The\npaper also shows our efforts in the digital identification of theMABs in 360°aspects by usingMatlab programme. This\nmethod of identification provides the single peak pattern of theMABs relative to the single flank pattern. Ithelpsusdevel-\nop the object-oriented software because this arithmetic has its capability in flexibility and speed, especially used in large\nareas. In addition, the paperbriefly introduces the potential capabilities of theMatlab programmewhich could resolve the\nproblem of identifying the upper and lower lmi its of theMABs. By comparing the twomethods, we indicate the relative\nmerits of the different techniques in digital identification and Tupu visualization, and show clearly their spatial scale for\nthe study regions. TheMABs from the digital identification have an agreementwith those from the field notes in theDaqing\nMountains, and the accuracy of theMABs from digital identification is decided by the original vegetationmaps. In other\nwords, the digital identification of theMABs is a processofdata integration and information diggingwithoutappendingun-\nnecessary information. Ifwe obtainedmore refined vegetationmaps as from higher resolution mi ages, we could enhance the\naccuracy of theMABs to amore satisfactory degree.\nThe study reveals that the digital identification of theMABs could enrich the sourcematerials ofmountain researches\nand enhance the digital comparison and analyticmethodsof theMABs. The digital identificationmodelsofaltitudinalbelts\ncould potentially help standardize the understanding ofaltitudinalbelts and their relation to various environmental factors.\nAnd italso could revealmore geographical information than everbefore and is an effective attempt forGeo-infoTupu prac-\ntices.