مقایسه گروه‌بندی و کیفیت مرزبندی‌های موروثی خاک با مدل‌های داده‌کاوی کمّی: مطالعه موردی بخش‌هایی از استان چهارمحال و بختیاری

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجو/دانشگاه شهرکرد

2 عضو هیئت علمی

3 گروه خاک، دانشکده کشاورزی، دانشگاه شهید باهنر کرمان

چکیده

کشف ارتباط بین خاک‌ها و گروه‌بندی آن‌ها بر اساس فاکتورهای مختلف، دربرگیرنده اهمیت بسزایی در زمینه‌های مختلف از جمله مدیریت اراضی و کشاورزی پایدار می‌باشد. این امر با ترسیم مدل‌های ذهنی و بر اساس فاکتورهای محیطی در قالب نقشه‌های سنتی خاک آغاز، و با استفاده از مفهوم فاصله و شباهت به کمک مدل‌های کمّی و ریاضی، ادامه یافته است. تحقیق حاضر به‌منظور مقایسه گروه‌بندی‌های در دسترس نقشه سنتی خاک با مدل‌های کمّی کلاسیک و مدرن صورت می‌گیرد. به این منظور، داده‌های موروثی خاک منطقه شهرکرد-بروجن با کمک الگوریتم‌های مختلف از جمله خوشه‌بندی سلسله مراتبی، میانگین‌های کا، درخت طبقه‌بندی و فاصله تاکسونومیکی گروه‌بندی شدند و نتایج با کلاس‌های خاک نقشه موروثی مورد مقایسه قرار گرفتند. نتایج بررسی‎ها در رزلوشن مکانی 90 متری نشان داد که گروه‌بندی‌های بدست آمده از درخت طبقه‌بندی، با کلاس‌های نقشه موروثی، بیشترین همخوانی را دارند. همچنین، الگوریتم‌های خوشه‌بندی سلسله مراتبی و میانگین‌های کا معمولی نیز به ترکیب گروهی مشابه با الگوی فیزیوگرافیک سنتی از نظر ویژگی‌های محیطی و مورفولوژیکی منجر شدند. آنالیز فاصله تاکسونومیکی با در نظر گرفتن همبستگی بین گروه‌ها و همچنین ویژگی‌های مورد بررسی، به بهترین ترکیب کلاس‌ها از نظر ویژگی‌های مختلف آن‌ها و همچنین بالاترین همبستگی درون کلاسی (522/0) و کمترین نسبت تغییر واریانس گروهی (915/0) منجر گردید. مقادیر کم آماره آنالیز تجزیه واریانس چندگانه (بین 001/0 در مدل سنتی تا 014/0 در مدل درختی)، نشان داد مدل‌های مورد بررسی بجز میانگین‌های کا، خاک‌ها را بطور مؤثری از یکدیگر تفکیک کرده‌اند. بطورکلی، استفاده از مدل‌های طبقه‌بندی عددی می‌توانند به نمایان کردن روابط کمّی بین خاک‌ها منجر شود. خاک‌شناس می‌تواند در نهایت ترکیب این گروه‌بندی‌ها را با استفاده از تجربه و آگاهی خویش از منطقه مورد بررسی در راستای دستیابی به گروه‌های یکنواخت‌تر از نظر ویژگی‌های مدیریتی و تاکسونومی آن‌ها تعدیل یا تصحیح کند.

کلیدواژه‌ها

عنوان مقاله [English]

Comparison of grouping and the quality of legacy soil map boundaries with numerical data mining models: a case study of some regions of Chaharmahal-va-Bakhtiari province

نویسندگان [English]

  • zahra rasaei 1
  • Jahangard Mohammadi 2
  • Azam Jafari 3
1
2 Soil Science Department, Agriculture faculty, Shahrekord University
3 Soil science department, Agriculture faculty, Shahid Bahonar University
چکیده [English]

Investigation of the the relationship between soils and grouping them based on different factors, plays an important role in different fields and aspects such as land management and sustainable agriculture. This launched by creating subjective or mental models and using environmental factors in the format of traditional soil maps. It afterward continued by relying on distance and similarity measurement through quantitative or mathematical models. This study aims to compare soil groups in soil maps with classical and modern models. For this purpose, legacy soil data of a map of Shahrekord-Borujen in Chaharmahal-va-Bakhtiari province is classified by using various algorithms: Heretical clustering, k-means, classification tree, and taxonomic distance. Soil groups gained from these numerical methods were then compared with soil groups in the legacy maps. Results of investigations in 90m spatial resolution showed that the classes from the decision tree were more in line with the legacy soil classes. The hierarchical clustering and k-means also resulted in group compositions similar to those of the legacy maps in terms of soil environmental and morphological characteristics that follow the traditional photographic units. The taxonomic distance led to the best combination of soil classes in the term of their traits with the highest within-class correlation (0.522) and the least within-group variance to between-group variance ratio. Low p < /em>-values in multivariate analysis of variance (MANOVA) between 0.001 in the traditional model up to 0.014 in the tree model showed that the models used in this study have effectively separated soils, except for K-means. Overall, findings show that using numerical classification models can discover the quantitative relations between soils. The surveyor can afterward modify the composition of soil classes considering his experience and knowledge of the study area in order to achieve more homogenous soil groups in terms of their management and taxonomic characteristics.

کلیدواژه‌ها [English]

  • Traditional soil map
  • Numerical classification
  • Boundary of map unit
  • Taxonomic distance
  • MANOVA

مراجع

Ashtekar J.M., and Owens P.R. 2013. Remembering knowledge: an expert knowledge based approach to digital soil mapping. Soil Horizons, 54(5): 1-6.
Bazaglia Filho O., Rizzo R., Lepsch I.F., do Prado H., Gomes F.H., Mazza J.A., and Demattê J.A.M. 2013. Comparison between detailed digital and conventional soil maps of an area with complex geology. Revista Brasileira de Ciência do Solo, 37: 1136–1148.
Boettinger J.L., Howell D.W., Moore A.C., Hartemink A.E., and Kienast-Brown S. 2010. Digital Soil Mapping: Bridging Research, Environmental Application, and Operation. Springer Science and Business Media, 433p.
Boettinger, J.L., Ramsey, R.D., Bodily, J.M., Cole, N.J., Kienast-Brown, S., Nield, S.J., Saunders, A.M. and Stum, A.K., 2008. Landsat spectral data for digital soil mapping. In: Hartemink A.E., McBratney A.B., Mendona-Santos M. (Eds.), Digital Soil Mapping with Limited Data. Springer, Dordrecht, pp. 193-202.
Bouma J., Montanarella L., and Evanylo G. 2019. The challenge for the soil science community to contribute to the implementation of the UN Sustainable Development Goals. Soil Use and Management, 35(4): 538-546.
Carré F., and Jacobson M. 2009. Numerical classification of soil profile data using distance metrics. Geoderma, 148: 336-345.
Conrad O., Bechtel B., Dietrich H., Fischer E., Gerlitz L., Wehberg J., Wichmann V., and Böhner J., 2015. System for Automated Geoscientific Analyses (SAGA) v. 2.1.4. Geoscientific Model Development, 8: 1991–2007.
Edokpayi A.A., Agho C.A., Adeh S.A., and Okpamen S.U. 2017. Comparison of the different hierarchical clustering techniques for the classification of soils under oil palm in Nigeria. International Journal of Basic Science and Technology, 3 (1): 37-46.
Everitt B.S., Landau S., Leese M., and Stahl M. 2011. Cluster Analysis (5th Ed.). Chichester, West Sussex, U.K., Wiley. 346p.
Hartigan J.A., and Wong M.A. 1979. Algorithm AS 136: A K-means clustering algorithm. Applied Statistics, 28: 100–108.
Heil J., Haring V., Marschner B., and Stumpe B. 2019. Advantages of fuzzy k-means over k-means clustering in the classification of diffuse reflectance soil spectra: A case study with West African soils. Geoderma, 337: 11-21.
Hughes P., McBratney A.B., Huang J., Minasny B., Micheli E., and Hempel J. 2017. Comparisons between USDA Soil Taxonomy and the Australian Soil Classification System I: Data harmonization, calculation of taxonomic distance and inter-taxa variation. Geoderma, 307:198-209.
Karaca S., Gülser F., and Selçuk R. 2018. Relationships between soil properties, topography and land use in the Van Lake Basin, Turkey. Eurasian Journal of Soil Science, 7 (2): 115-120.
Kaufman L., and Rousseeuw P.J. 2005. Finding Groups in Data: An Introduction to Cluster Analysis. Wiley, Hoboken, New York. 342p.
Kempen B., Brus D.J., Stoorvogel J.J., Heuvelink G.B.M., and de Vries F. 2012. Efficiency comparison of conventional and digital soil mapping for updating soil maps. Soil Science Society of America Journal, 76: 2097–2115.
Kumar D.A., and Kannathasan N. 2013. A study and characterization of chemical properties of soil surface data using K-means algorithm. In: International Conference on Pattern Recognition, Informatics and Mobile Engineering, pp. 264-270.
Mahler P.J. 1970. Manual of multipurpose land classification. Joint project with Food and Agriculture Organization (FAO-Undp). Publication No. 212. Ministry of Agriculture, Soil Institute of Iran.
Michéli E., Láng V., Owens P.R., McBratney A.B., and Hempel J. 2016. Testing the pedometric evaluation of taxonomic units on soil taxonomy-a step in advancing towards a universal soil classification system. Geoderma, 264: 340-349.
Mohammadi J. 2013. Pedomining, Vol. 3, Postmodern Statistics. Pelk Press. Tehran, Iran. 594p. (In Persian)
Mohammadi M. 1986. Semi-detailed soil studies report Chaharmahal-Va-Bakhtiari province (Shahrekord and Borujen areas). Iranian Soil and Water Research Institute. No. 696. Tehran, Iran. (In Persian)
Nield S.J., Boettinger J.L., and Ramsey R.D. 2007. Digitally mapping gypsic and natric soil areas using Landsat ETM data. Soil Science Society of America Journal, 71: 245-252.
Ovalles F.A., and Collins M.E. 1986. Soil-landscape relationships and soil variability in north central Florida. Soil Science Society of America Journal, 50(2): 401-408.
Palepu R.B., and Muley R.R. 2017. An Analysis of Agricultural Soils by using Data Mining Techniques. International Journal of Engineering Science and Computing, 7(10): 15167-15177.
Qishlaqi A., Moore F., and Forghani G. 2009. Characterization of metal pollution in soils under two landuse patterns in the Angouran region, NW Iran; a study based on multivariate data analysis. Journal of Hazardous Materials, 172: 374–384.
Quinlan J.R. 1986. Induction of Decision Trees. Machine Learning, 1(1): 81–106.
R Development Core Team. 2017. R: A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria.
Ribeiro M.V., Cunha L.M.S., Camargo H.A., and Rodrigues L.H.A. 2014. Applying a fuzzy decision tree approach to soil classification. In: International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems. pp. 87-96.
Rouse Jr J.W., Haas R.H., Schell J.A., and Deering D.W. 1974. Monitoring Vegetation Systems in the Great Plains with Erts. NASA Special Publication, 351: 309-317.
Soil Survey Staff. 2014. Keys to Soil Taxonomy (12th Ed.). USDA-Natural Resources Conservation Service, Washington DC. 372p.
Toomanian N., and Esfandiarpour Boroujeni I. 2017. Outcomes of applying a geopedologic approach to soil survey in Iran. Desert, 22-2: 239-247.
Uribeetxebarria A., Arnó J., Escolà A., and Martínez-Casasnovas J.A. 2018a. Apparent electrical conductivity and multivariate analysis of soil properties to assess soil constraints in orchards affected by previous parcelling. Geoderma, 319: 185–193.
Van Huyssteen C.W., Turner D.P., and Le Roux P.A.L. 2013. Principles of soil classification and the future of the South African system. South African Journal of Plant and Soil, 30(1): 23-32.
Yang L., Jiao Y., Fahmy S., Zhu A-X., Hann S., Burt J.E., and Qi F. 2011. Updating Conventional Soil Maps through Digital Soil Mapping. Soil Science Society of America Journal, 75: 1044–1053.
Zinck J.A., Metternicht G., Bocco G., and Del Valle H.F. (Eds.). 2016. Geopedology: An Integration of Geomorphology and Pedology for Soil and Landscape Studies (1st Ed.). Springer, New York, 550p.
Zirnea S., Lazar I., Saha Foudjo B-U., Vasilache T., and Lazar G. 2013. Cluster Analysis Based of Geochemical Properties of Phosphogypsum Dump Located Near Bacau City in Romania. APCBEE Procedia, 5: 317-322.

فایل ها

سابقه مقاله

  • تاریخ دریافت: 07 مهر 1398
  • تاریخ بازنگری: 27 دی 1399
  • تاریخ پذیرش: 23 دی 1398

هم رسانی

ارجاع به این مقاله

آمار