Digital mapping of soil organic carbon using auxiliary data and machine learning models in Badr watershed, south of Qorve city, Kurdistan province.

Document Type : Complete scientific research article

Authors

1 Soil and Water Research Expert, Agricultural and Natural Resources Research and Training Center of Tehran Province, Agricultural Research, Training and Extension Organization (TAT), Tehran, Iran.

2 Assistant Professor of Geology, Faculty Member of the Research and Education Center for Agriculture and Natural Resources of Tehran Province, Agricultural Research, Education and Extension Organization (TAT), Tehran, Iran.

Abstract

Soil organic carbon is considered as a key factor in the stability of soil fertility and soil ecosystem services. Soil organic carbon is also included in the United Nations Environment Program as one of the important environmental issues and challenges on a global scale. Studies in Iran show that, on average, for each gram of organic carbon in a kilogram of soil, the yield of wheat increases by an average of 286 kilograms per hectare. Also, knowledge of soil organic carbon changes is one of the main components in soil quality evaluation. The goals of this study are digital mapping of soil carbon and identification of the effects of environmental characteristics on predictions of soil organic carbon, Therefore, the present study was conducted with the aim of digital mapping of soil organic carbon using environmental auxiliary variables and predictive models and introduction of the best models in the Badr watershed in the south of Qorve city. In order to conduct this research in the first stage, auxiliary data such as Landsat 8 satellite images and a digital elevation model with a spatial resolution of 10 meters were obtained from the country's mapping organization. The geological map of Qorveh was prepared from the geological site of the country, and the geological map of the Badr watershed was extracted from it and digitized in the environment of the geographic information system. The geomorphological map was drawn using the geological map and based on the zinc geopedology method in the environment of the geographic information system. In the second stage, the location of the observation points was determined, soil identification was done in the desert, sampling was done from different soil layers, and physical, chemical and mineralogical measurements of the soils were done and the soils were classified. . In the third stage, modeling was done, digital maps of soil classes and characteristics were prepared and the models were evaluated. To conduct this study, based on the Latin super cube technique, 125 outcrops were selected and excavated in the study area. After air-drying in the laboratory environment, the soil samples were pounded and passed through a 2 mm sieve. In the first case, artificial neural network models, decision tree analysis and linear multivariate regression were used for prediction. The results of these predictions were evaluated using the k-fold 5 random evaluation method. In the second step, decision tree analysis, artificial neural network, nearest neighbor and random forest models were used for prediction. Also, in order to combine the results of the models in this case, the mixed linear multivariate regression method was used. By using the spatial k-fold 10 evaluation method, the prediction results were evaluated. The results showed that among the models used for predicting soil organic matter, the multivariate linear regression (MLR) model with the coefficient of determination of 0.637 and the square root of the mean square error of 0.232 had the highest accuracy for prediction. The lowest prediction accuracy is assigned to the K nearest neighbor (KNN) model. Meanwhile, by using the k-fold 5 random validation method, among the models of Artificial Neural Network (ANN), Decision Tree Analysis (DTA), Multi Linear Regression, MLR) and K Nearest Neighbor, the K Nearest Neighbor (KNN) model with a coefficient of determination of 0.9906 and a square root mean square error of 0.0361 has the most accuracy for predicting the amount of organic carbon. It goes without saying that due to the partiality of the location k-fold 10 validation method, using this method is preferable to the random k-fold 5 validation method.

Keywords

Main Subjects

References

 1.McBratney, A. B., Santos, M. M., & Minasny, B. (2003). On digital soil mapping. Geoderma, 117(1-2), 3-52. doi.org/10.1016/S0016-7061(03)00223-4.
2.Minasny, B., McBratney, A. B., Malone, B. P., & Wheeler, I. (2013). Digital mapping of soil carbon. Advances in agronomy, 118, 1-47. doi.org/10.1016/ B978-0-12-405942-9.00001-3.
3.Hengl, T., Heuvelink, G. B., Kempen, B., Leenaars, J. G., Walsh, M. G., Shepherd, K. D., ... & Tondoh, J. E. (2015). Mapping soil properties of Africa at 250 m resolution: Random forests significantly improve current predictions. PloS one, 10(6), e0125814. doi.org/ 10.1371/journal.pone.0125814.
4.Powers, J. S., Corre, M. D., Twine, T. E., & Veldkamp, E. (2011). Geographic bias of field observations of soil carbon stocks with tropical land-use changes precludes spatial extrapolation. Proceedings of the National Academy of Sciences, 108(15), 6318-6322. doi.org/10.1073/ pnas.1016774108.
5.Grüneberg, E., Schöning, I., Hessenmöller, D., Schulze, E. D., & Weisser, W. W. (2013). Organic layer and clay content control soil organic carbon stocks in density fractions of differently managed German beech forests. Forest Ecology and Management, 303, 1-10. DOI:10.1016/J.FORECO.2013.03.014.
6.Mayer, L. M. (1994). Relationships between mineral surfaces and organic carbon concentrations in soils and sediments. Chemical Geology, 114(3-4), 347-363. doi.org/10.1016/0009-2541(94)90063-9.
7.Bui, E., Henderson, B., & Viergever, K. (2009). Using knowledge discovery with data mining from the Australian Soil Resource Information System database to inform soil carbon mapping in Australia. Global biogeochemical cycles, 23(4). doi.org/10.1029/2009GB003506.
8.Yang, R. M., Zhang, G. L., Yang, F., Zhi, J. J., Yang, F., Liu, F., ... & Li, D. C. (2016). Precise estimation of soil organic carbon stocks in the northeast Tibetan Plateau. Scientific reports, 6(1), 21842. DOI: 10.1038/srep21842.
9.Fissore, C., Dalzell, B. J., Berhe, A. A., Voegtle, M., Evans, M., & Wu, A. (2017). Influence of topography on soil organic carbon dynamics in a Southern California grassland. Catena, 149. 140-149. DOI:10.1016/j.catena.2016.09.016.
10.Nyssen, J., Temesgen, H., Lemenih, M., Zenebe, A., Haregeweyn, N., & Haile, M. (2008). Spatial and temporal variation of soil organic carbon stocks in a lake retreat area of the Ethiopian Rift Valley. Geoderma, 146(1-2), 261-268. DOI:10.1016/j.geoderma.2008.06.007.
11.Oueslati, I., Allamano, P., Bonifacio, E., & Claps, P. (2013). Vegetation and topographic control on spatial variability of soil organic carbon. Pedosphere, 23(1), 48-58. DOI:10.1016/S1002-0160 (12)60079-4.
12.Yang, R. M., Zhang, G. L., Liu, F., Lu, Y. Y., Yang, F., Yang, F., ... & Li, D. C. (2016). Comparison of boosted regression tree and random forest models for mapping topsoil organic carbon concentration in an alpine ecosystem. Ecological indicators, 60, 870-878. DOI:10.1016/j.ecolind. 2015.08.036.
13.Hounkpatin, O. K., de Hipt, F. O., Bossa, A. Y., Welp, G., & Amelung, W. (2018). Soil organic carbon stocks and their determining factors in the Dano catchment (Southwest Burkina Faso). Catena, 166, 298-309. DOI:10.1016/ j.catena.2018.04.013.
14.Ließ, M., Schmidt, J., & Glaser, B. (2016). Improving the spatial prediction of soil organic carbon stocks in a complex tropical mountain landscape by methodological specifications in machine learning approaches. PLoS One,
11(4), e0153673. doi.org/10.1371/ journal.pone.0153673.
15.Wang, B., Waters, C., Orgill, S., Gray, J., Cowie, A., Clark, A., & Li Liu, D. (2018). High resolution mapping of soil organic carbon stocks using remote sensing variables in the semi-arid rangelands of eastern Australia. Science of the Total Environment, 630, 367-378. doi.org/10.1016/j.scitotenv.2018.02.204.
16.Adhikari, K., Hartemink, A. E., Minasny, B., Bou Kheir, R., Greve, M. B., & Greve, M. H. (2014). Digital mapping of soil organic carbon contents and stocks in Denmark. PloS one,
9(8), e105519. doi.org/10.1371/journal. pone.0105519.
17.Gray, J. M., Bishop, T. F., & Yang, X. (2015). Pragmatic models for the prediction and digital mapping of soil properties in eastern Australia. Soil Research, 53(1), 24-42. DOI:10.1071/ SR13306.
18.Rossel, R. V., Brus, D. J., Lobsey, C., Shi, Z., & McLachlan, G. (2016). Baseline estimates of soil organic carbon by proximal sensing: Comparing design-based, model-assisted and model-based inference. Geoderma, 265, 152-163. DOI:10.1016/j.geoderma.2015.11.016.
19.Ottoy, S., De Vos, B., Sindayihebura, A., Hermy, M., & Van Orshoven, J. (2017). Assessing soil organic carbon stocks under current and potential forest cover using digital soil mapping
and spatial generalisation. Ecological indicators, 77, 139-150. DOI:10.1016/ j.ecolind.2017.02.010.
20.Minasny, B., Setiawan, B. I., Saptomo, S. K., & McBratney, A. B. (2018). Open digital mapping as a cost-effective method for mapping peat thickness and assessing the carbon stock of tropical peatlands. Geoderma, 313, 25-40. DOI:10.1016/j.geoderma.2017.10.018.
21.Bonfatti, B. R., Hartemink, A. E., Giasson, E., Tornquist, C. G., & Adhikari, K. (2016). Digital mapping of soil carbon in a viticultural region of Southern Brazil. Geoderma, 261, 204-221. DOI:10.1016/j.geoderma.2015.07.016.
22.Gamble, J. D., Feyereisen, G. W., Papiernik, S. K., Wente, C., & Baker, J. (2017). Regression‐Kriged Soil Organic Carbon Stock Changes in Manured Corn Silage–Alfalfa Production Systems. Soil Science Society of America Journal, 81(6), 1557-1566. DOI:10.2136/sssaj 2017.04.0138.
23.Batty, M., & Torrens, P. M. (2001). Modelling complexity: the limits to prediction. Cybergeo: European Journal of Geography. https://doi.org/10.4000/ cybergeo.1035.
24.Brungard, C. W., Boettinger, J. L., Duniway, M. C., Wills, S. A., & Edwards Jr, T. C. (2015). Machine learning for predicting soil classes in three semi-arid landscapes. Geoderma, 239, 68-83. DOI:10.1016/j.geoderma. 2014.09.019.
25.Ballabio, C. (2009). Spatial prediction of soil properties in temperate mountain regions using support vector regression. Geoderma, 151(3-4), 338-350. doi.org/ 10.1016/j.geoderma.2009.04.022.
26.Hounkpatin, O. K., de Hipt, F. O., Bossa, A. Y., Welp, G., & Amelung, W. (2018). Soil organic carbon stocks and their determining factors in the Dano catchment (Southwest Burkina Faso). Catena, 166, 298-309. DOI:10.1016/ j.catena.2018.04.013.
27.Stevens, A., Nocita, M., Tóth, G., Montanarella, L., & van Wesemael, B. (2013). Prediction of soil organic carbon at the European scale by visible and near infrared reflectance spectroscopy. PloS one, 8(6), e66409. doi.org/10.1371/ journal.pone.0066409.
28.Vašát, R., Kodešová, R., Borůvka, L., Jakšík, O., Klement, A., & Brodský, L. (2017). Combining reflectance spectroscopy and the digital elevation model for soil oxidizable carbon estimation. Geoderma, 303, 133-142. doi.org/10.1016/j.geoderma.2017.05.018.
29.Minasny, B., McBratney, A. B., Mendonça-Santos, M. D. L., Odeh, I. O. A., & Guyon, B. (2006). Prediction and digital mapping of soil carbon storage in the Lower Namoi Valley. Soil Research, 44(3), 233-244. DOI:10.1071/SR05136.
30.Mishra, U., Lal, R., Slater, B., Calhoun, F., Liu, D., & Van Meirvenne, M. (2009). Predicting soil organic carbon stock using profile depth distribution functions and ordinary kriging. Soil Science Society of America Journal, 73(2), 614-621. doi.org/10.2136/sssaj 2007.0410.
31.Bishop, T. F. A., McBratney, A. B., & Laslett, G. M. (1999). Modelling soil attribute depth functions with equal-area quadratic smoothing splines. Geoderma, 91(1-2), 27-45. DOI:10.1016/S0016-7061(99)00003-8.
32.Adhikari, K., Kheir, R. B., Greve, M. B., Bøcher, P. K., Malone, B. P., Minasny, B., ... & Greve, M. H. (2013). High‐resolution 3‐D mapping of soil texture in Denmark. Soil Science Society of America Journal, 77(3), 860-876. DOI:10.2136/sssaj2012.0275.
33.Mulder, V. L., Lacoste, M., Richer-de-Forges, A. C., Martin, M. P., & Arrouays, D. (2016). National versus global modelling the 3D distribution of soil organic carbon in mainland France. Geoderma, 263, 16-34. DOI: 10.1016/ j.geoderma.2015.08.035.
34.Fontaine, S., Barot, S., Barré, P., Bdioui, N., Mary, B., & Rumpel, C. (2007). Stability of organic carbon in deep soil layers controlled by fresh carbon supply. Nature, 450(7167), 277-280. DOI:10. 1038/nature06275.
35.Meersmans, J., van Wesemael, B., De Ridder, F. A., & Van Molle, M. (2009). Modelling the three-dimensional spatial distribution of soil organic carbon (SOC) at the regional scale (Flanders, Belgium). Geoderma, 152(1-2), 43-52. DOI:10.1016/j.geoderma.2009.05.015.
36.Pan, L., & Politis, D. N. (2016). Bootstrap prediction intervals for linear, nonlinear and nonparametric autoregressions. Journal of Statistical Planning and Inference, 177, 1-27. doi.org/10.1016/j.jspi.2014.10.003.
37.Shrestha, D. L., & Solomatine, D. P. (2006). Machine learning approaches for estimation of prediction interval for the model output. Neural networks, 19(2), 225-235. DOI:10.1016/j.neunet. 2006.01.012.
38.Solomatine, D. P., & Shrestha, D. L. (2009). A novel method to estimate model uncertainty using machine learning techniques. Water Resources Research, 45(12). doi.org/10.1029/ 2008WR006839.
39.Malone, B. P., McBratney, A. B., & Minasny, B. (2011). Empirical estimates of uncertainty for mapping continuous depth functions of soil attributes. Geoderma, 160(3-4), 614-626. DOI:10.1016/j.geoderma.2010.11.013.
40.Arrouays, D., Grundy, M. G., Hartemink, A. E., Hempel, J. W., Heuvelink, G. B., Hong, S. Y., ... & Zhang, G. L. (2014). GlobalSoilMap: Toward a fine-resolution global grid of soil properties. Advances in agronomy, 125, 93-134. DOI: 10.1016/B978-0-12-800137-0.00003-0.
41.Hariri, A. (1995). An attitude on the origin of a group of different rocks in the Qorve area. Master's thesis, Shahid Beheshti University, Tehran. [In Persian]
42.Hosseini, M. (1996). Description of Geological Map 1:100000 Quarter Corners (Map Attachment), Geological and Mineral Exploration Organization of the country. [In Persian]
43.Zinck, J. A. (1989). Physiography and soils. Lecture notes for soil students. Soil Science Division. Soil survey courses subject matter: K6 ITC, Enschede, The Netherlands.
44.Schoeneberger, P. J., Wysocki, D. A., & Benham, E. C. (Eds.). (2012). Field book for describing and sampling soils. Government Printing Office.
45.Walkley, A., & Black, I. A. (1934). An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil science, 37(1), 29-38. doi.org/10.1097/ 00010694-193401000-00003.
46.Givi, J. (1997). Qualitative assessment of land suitability for agricultural and garden plants, Soil and Water Research Institute. Technical magazine 1015, 100 p. [In Persian]
47.Valavi, R., Elith, J., Lahoz-Monfort, J. J., & Guillera-Arroita, G. (2018). blockCV: An r package for generating spatially or environmentally separated folds for k-fold cross-validation of species distribution models. Biorxiv, 357798. DOI:10.1111/2041-210X.13107.
48.Balan, B., Mohaghegh, S., & Ameri, S. (1995, September). State-of-the-art in permeability determination from well log data: Part 1-A comparative study, model development. In SPE Eastern Regional Meeting (pp. SPE-30978). SPE. DOI:10.2118/30979-MS.
49.Matinfar, H. R., Mghsodi, Z., Mossavi, S. R., & Jalali, M. (2021). Evaluation of Machine Learning Methods in
Digital Mapping of Soil Organic Carbon (part of Khorramabad Plain). Journal of Water and Soil Science, 24 (4), 327-342. 10.47176/jwss.24.4.8411. (In Farsi)
50.Salehi, M. H., & Khademi, H. (2017). Basics of soil mapping. Isfahan University Jihad Publications. 210 p. [In Persian]
51.Behrens, T., Förster, H., Scholten, T., Steinrücken, U., Spies, E. D., & Goldschmitt, M. (2005). Digital soil mapping using artificial neural networks. Journal of plant nutrition and soil science, 168(1), 21-33. https://doi.org/10.1002/jpln.200421414.
52.Angiulli, F. (2005, August). Fast condensed nearest neighbor rule. In Proceedings of the 22nd international conference on Machine learning (pp. 25-32).
53.Nemes, A., Rawls, W. J., & Pachepsky, Y. A. (2006). Use of the nonparametric nearest neighbor approach to estimate soil hydraulic properties. Soil Science Society of America Journal, 70(2), 327-336. DOI:10.2136/sssaj 2005.0128.
54.Zolfaghari, A. A., Tirgar Soltani, M. T., Afshart, T., & Sarmadian, F. (2013). Comparison of K-nearest neighbor and artificial neural network methods for predicting cation exchange capacity of soil. Journal of Soil Management and Sustainable Production, 3(1), 77-94. https://dor.isc.ac/dor/20.1001.1.23221267.1392.3.1.5.2. [In Persian]
55.Ayoubi, S., Taghizadeh, R., Namazi, Z., Zulfiqari, A., & Rustaee Sadrabadi, F., (2015). Comparison of k-nearest neighbor and artificial neural network methods for digital zoning of soil salinity in Afzal Ardakan well area. Journal of Water and Soil Science. https:// civilica.com/ doc/ 1201298. http://dx.doi.org/10.18869/acadpub.jstnar.20.76.59. [In Persian]
56.Scull, P., Franklin, J., & Chadwick, O. A. (2005). The application of classification tree analysis to soil type prediction in a desert landscape. Ecological modelling, 181(1), 1-15. DOI:10.1016/j.ecolmodel.2004.06.036.
57.Moore, I. D., Grayson, R. B., & Ladson, A. R. (1991). Digital terrain modelling: a review of hydrological, geomorphological, and biological applications. Hydrological processes, 5(1), 3-30. https://doi.org/10.1002/ hyp.3360050103.
58.Breiman, L. (2001). Random forests. Machine learning, 45, 5-32. http://dx. doi.org/10.1023/A:1010933404324.
59.Heung, B., Bulmer, C. E., & Schmidt, M. G. (2014). Predictive soil parent material mapping at a regional-scale: A Random Forest approach. Geoderma, 214, 141-154. DOI:10.1016/j.geoderma. 2013.09.016.
60.Wilding, L., & Drees, L. R. (1983). Spatial variability and pedology. In Developments in Soil Science
(Vol. 11, pp. 83-116). Elsevier.
ttps:// doi.org/ 10.1016/ S0166-2481 (08)70599-3.
61.Yoo, K., Amundson, R., Heimsath, A. M., & Dietrich, W. E. (2006). Spatial patterns of soil organic carbon on hillslopes: Integrating geomorphic processes and the biological C cycle. Geoderma, 130(1-2), 47-65. https:// doi. org/10.1016/j.geoderma.2005.01.008.
62.Bangroo, S. A., Najar, G. R., & Rasool, A. (2017). Effect of altitude and aspect on soil organic carbon and nitrogen stocks in the Himalayan Mawer Forest Range. Catena, 158, 63-68. DOI:10.1016/j.catena.2017.06.017.
63.Mirzaee, S., Ghorbani-Dashtaki, S., Mohammadi, J., Asadi, H., & Asadzadeh, F. (2016). Spatial variability of soil organic matter using remote sensing data. Catena, 145, 118-127. DOI:10.1016/j.catena.2016.05.023.
Journal of Soil Management and Sustainable Production
Volume 15, Issue 2
June 2025
Pages 1-22

Files

Share

How to cite

Statistics