تهیۀ نقشه‌های سه‌بعدی اجزای بافت خاک با تلفیق الگوریتم جنگل رگرسیونی چندکی و تابع عمق اسپیلاین در استان گلستان

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه علوم خاک-دانشکده ی مهندسی آب و خاک-دانشگاه علوم کشاورزی و منابع طبیعی گرگان

2 گروه علوم خاک، دانشکده مهندسی آب و خاک، دانشگاه علوم کشاورزی و منابع طبیعی گرگان، ایران.

3 تحقیقات خاک و آب، مرکز تحقیقات و آموزش کشاورزی و منابع طبیعی گلستان، گرگان

4 گروه علوم خاک، دانشکده مهندسی آب و خاک، دانشگاه علوم کشاورزی و منابع طبیعی گرگان. ایران.

چکیده

امروزه نیاز روزافزونی به اطلاعات مکانی پیوسته و کمی خاک در راستای مدل‌سازی و مدیریت محیطی، به‌ویژه در مقیاس ملی وجود دارد. این مطالعه با هدف پیش‌بینی نسبت اندازه ذرات خاک (PSF) در بخشی از اراضی استان گلستان با استفاده از تلفیق مدل جنگل رگرسیونی چندکی (QRF) و تابع اسپیلاین انجام شد. تابع عمق اسپیلاین با مساحت برابر برای تخمین PSFs در پنج عمق خاک (0-25، 25-50، 50-75، 75-100، و 100-125 سانتی‌متر) به داده‌های 105 خاکرخ از بانک اطلاعات دانشگاه علوم کشاورزی و منابع طبیعی گرگان برازش داده شد. متغیرهای کمکی اولیه در این تحقیق شامل 22 متغیر محیطی مشتق شده از DEM، 15 شاخص سنجش از دور از ماهواره لندست هفت سنجنده ETM+، نقشه‌های عمق ایستابی (پیزومتری) و بارندگی بودند. بر اساس روش تجزیه مؤلفه‌های اصلی (PCA)، 15متغیر انتخاب و وارد فرآیند مدل‌سازی اجزای بافت خاک (رس، سیلت و شن) شدند. عملکرد مدل QRF با استفاده از آماره‌های ضریب تبیین (R2)، ریشه میانگین مربعات خطا (RMSE)، و قدر مطلق میانگین خطا (MAE) مورد ارزیابی قرار گرفت. نتایج نشان داد میزان ضریب تببین برای رس، سیلت، و شن در عمق‌های مختلف به ترتیب از 12/0 تا 22/0، 07/0 تا 30/0، و 07/0 تا 28/0 متغیر بود. همچنین اهمیت نسبی متغیر‌های محیطی نشان داد بارندگی (میانگین سی‌ساله)، عمق ایستابی (میانگین ده‌ساله)، B3/B7 و شاخص عمق دره، مهمترین پارامترهای کنترل‌کنندۀ اجزای بافت خاک در تحقیق حاضر بودند. به منظور بهبود عملکرد مدل و نتایج اعتبارسنجی نیاز به پرداختن به برخی عدم قطعیت‌های ساختاری در این مطالعه وجود دارد.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Preparation of three-dimensional maps of soil particle size fractions by combining quantile regression forest algorithm and spline depth function in Golestan Province

نویسندگان [English]

  • Maryam Emami 1
  • Farhad Khormali 2
  • Mohammad reza Pahlavan Rad 3
  • Soheila Ebrahimi 4
1 Department of Soil Science, Faculty of water and soil Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran.
2 Corresponding Author, Department of Soil Science, Faculty of water and soil Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran.
3 Soil and Water Research Department, Golestan Agricultural and Natural Resources Research and Education Center, AREEO, Gorgan, Iran.
4 Department of Soil Science, Faculty of water and soil Engineering, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran
چکیده [English]

There is an increasing need for continuous spatial and quantitative soil information for environmental modeling and management, especially at the national scale. This study was conducted to predict the soil particle size fraction (PSF) using the combination of quantile regression forest model (QRF) and spline function in a part of Golestan province. An equal area spline equation was fitted to the data of 105 soil profiles from the database of the Gorgan University of Agricultural Sciences and Natural Resources for estimating PSFs at five soil depths (0-25, 25-50, 50-75, 75-100, and 100-125 cm). The primary auxiliary variables in this research included 22 environmental variables derived from DEM, 15 remote sensing indicators obtained from the Landsat 7 ETM+ images, rainfall and piezometric maps. Based on principal component analysis (PCA), 15 variables were selected and entered into the modeling process of soil texture components (clay, sand, and silt). The efficiency of the quantile regression forest model was evaluated using the coefficient of determination (R2), the root mean squared error (RMSE), and the mean absolute error (MAE). The results indicated that the coefficient of determination for clay, silt, and sand at different depths varied from 0/12 to 0/22, 0/07 to 0/30, and 0/07 to 0/28, respectively. Also, the relative importance of environmental variables showed that rainfall (thirty-year average), piezometry (ten-year average), B3/B7, and valley depth were the most important factors in predicting soil texture components. To improve model performance and validation results, some structural uncertainties in this study should be addressed.

کلیدواژه‌ها [English]

  • Principal component analysis (PCA)
  • Quantile regression forest (QRF)
  • Soil particle size fraction (PFS)
  • Spline

Preparation of three-dimensional maps of soil particle size fractions by combining quantile regression forest algorithm and spline depth function in Golestan Province

EXTENDED ABSTRACT

Introduction

Soil texture, which includes three components of sand, clay, and silt, is one of the important physical characteristics of soil that strongly affects many other soil properties, such as water retention curve, fertility, drainage, organic carbon content, and porosity. Knowledge of the spatial variability of soil properties, including soil texture, is essential in precision agriculture because any change in the spatial distribution of physical and chemical soil properties causes changes in crop yield. Digital soil mapping (DSM) techniques apply the analysis of soil spatial relationships, terrain features, and remote sensing images to environmental modeling. Soil properties in general vary continuously with depth in a soil profile. Spline functions are very efficient in modeling soil attribute depth functions. Combining spline functions and digital mapping is a suitable approach for the 3D modeling soil properties. In recent years, many digital mapping studies have been conducted in Iran. However, to our knowledge, no study has been conducted on depth functions to investigate soil properties in Golestan Province. Therefore, this study aims to estimate soil texture components using the spline function at the target depths, spatial modeling of soil texture components with the quantitative regression forest model (QRF), and evaluate the contribution of environmental variables for modeling in a part of the north of Golestan Province.

Materials and Methods

The data from 105 profiles of the soil information bank of the University of Agricultural Sciences, which were collected between 2013 and 2015, were used in this research. Equal-area quadratic smoothing splines were used to describe the vertical variation of soil particle size fractions. The next step was to select the most useful of the 39 predictor ancillary variables to reduce the dimensionality and allow the QRF algorithm to operate more effectively. Here, principal component analysis (PCA) was used to rank the relevance of auxiliary variables. 15 auxiliary variables including rainfall, piezometry, carbonate index, clay index, B3/B7, NDVI, MrRTF, MrVBF, morphometric feature, landform, valley depth, flow width, surface wetness index, annual insolation, and slope were selected to predict soil texture components. Twenty-fold cross-validation was used to evaluate the performances of the QRF algorithm. To determine the accuracy of the model, three different criteria were used based on the coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE).

Results and Discussion

The validation results of the QRF learning algorithm in predicting the spatial changes of soil texture components showed that the coefficient of determination for clay ranged from 0/12 to 0/22, sand from 0/07 to 0/28, and silt from 0/07 to 0/30. According to the findings, auxiliary variables contributed differently to predicting soil texture components at different depths. Based on the relative importance of the variables, the distribution of clay and silt was more affected by rainfall and groundwater depth. The higher rainfall in the southern part of the study area increased silt weathering and clay accumulation. In the northern lowlands of the study area, impermeable clay layers were formed due to the accumulation of fine-textured sediments from the Atrak and Gorganrud Rivers, which affected the piezometric factor. However, piezometric in the southern regions is a measure of the groundwater depth, and its spatial distribution pattern is consistent with the spatial distribution pattern of clay in the south. The maximum amount of silt in the study area was found in the middle regions, mainly due to the floodplain and sedimentary plain of the Gorganrud River. The distribution of sand was more influenced by the two environmental factors of valley depth and B3/B7. The pattern of spatial changes of these two indicators is in line with the spatial distribution of sand.

Conclusions

Based on the findings of this research, the amounts of clay, sand, and silt had a strong relationship with rainfall and groundwater levels in the study area. The quantile regression forest algorithm showed poor performance in predicting soil particle size in the study area. Data augmentation is effective in reducing uncertainty and enhancing model accuracy.

Ahmadi, M., Marvati, A., Nojavan, M. R., & Ghasemi, A. )2015(. Investigating the relationship between soil texture, vegetation, and groundwater table depth in different geomorphic levels of the Chah Afzal desert area in Yazd province. The third Sustainable Agriculture and Natural Resources Conference. 1-8. (In Persian)
Akpa, S.I., Odeh, I.O., Bishop, T.F, & Hartemink, A.E. (2014). Digital mapping of soil particle‐size fractions for Nigeria. Soil Science Society of America Journal78(6), 1953-1966.
Amirian Chekan, A., Taghizadeh Mehrjerdi, R., Sarmadian, F., &Heidary, A. (2017). Three-dimensional mapping of soil texture using spline depth functions and artificial neural networks. Iranian Journal of Soil and Water Research, 48(1),113-123. (In Persian)
Bishop, T.F.A., McBratney, A.B. & Laslett, G.M. (1999). Modelling soil attribute depth functions with equal-area quadratic smoothing splines. Geoderma91(1-2), 27-45.
CarvalhoJunior, W.D., Chagas, C.D.S., FernandesFilho, E.I., Vieira, C.A.O., Schaefer, C.E.G., Bhering, S.B. & Francelino, M.R. (2011). Digital soilscape mapping of tropical hillslope areas by neural networks. Scientia Agricola68, 691-696.
Chamizo, S., Canton, Y., Lázaro, R., & Domingo, F. (2013). The role of biological soil crusts in soil moisture dynamics in two semiarid ecosystems with contrasting soil textures. Journal of Hydrology, 489, 74-84.
Dharumarajan, S., Kalaiselvi, B., Suputhra, A., Lalitha, M., Hegde, R., Singh, S.K. & Lagacherie, P. (2020). Digital soil mapping of key GlobalSoilMap properties in Northern Karnataka Plateau. Geoderma Regional20, e00250.
Di Fusco, E., Lauriola, I., Verdone, R., Di Federico, V., & Ciriello, V. (2018). Impact of uncertainty in soil texture parameters on estimation of soil moisture through radio waves transmission. Advances in Water Resources, 122, 131-138.
Florinsky, I.V., Eilers, R.G., Manning, G.R., Fuller, L.G. (2002). Prediction of soil properties by digital terrain modelling. Environmental Modelling & Software, 17, 295–311.
Gastaldi, G., Minasny, B. & McBratney, A.B. (2012). Mapping the occurrence and thickness of soil horizons within soil profiles. In Digital soil assessments and beyond (pp. 145-148). CRC Press/Balkema London.
Ge, X., Ding, J., Teng, D., Wang, J., Huo, T., Jin, X., Wang, J., He, B., & Han, L. (2022). Updated soil salinity with fine spatial resolution and high accuracy: The synergy of Sentinel-2 MSI, environmental covariates and hybrid machine learning approaches. Catena212, 106054.
Greve, M.H., Kheir, R.B., Greve, M.B., & Bøcher, P.K. (2012). Using digital elevation models as an environmental predictor for soil clay contents. Soil Science Society of America Journal76(6), 2116-2127.
Hengl, T., Heuvelink, G.B., Rossiter, D.G. (2007). About regression-kriging: from equations to case studies. Computers and Geosciences 33, 1301–1315.
Hengl, T., Mendes de Jesus, J., Heuvelink, G.B., Ruiperez Gonzalez, M., Kilibarda, M., Blagotić, A., Shangguan, W., Wright, M.N., Geng, X., Bauer-Marschallinger, B., & Guevara, M.A. (2017). SoilGrids250m: Global gridded soil information based on machine learning. PLoS one12(2), e0169748.
Jena, R.K., Moharana, P.C., Dharumarajan, S., Sharma, G.K., Ray, P., Deb Roy, P., Ghosh, D., Das, B., Alsuhaibani, A.M., Gaber, A., & Hossain, A. (2023). Spatial Prediction of Soil Particle-Size Fractions Using Digital Soil Mapping in the North Eastern Region of India. Land, 12(7), 1295.
Kempen, B., Brus, D.J, & Stoorvogel, J.J. (2011). Three-dimensional mapping of soil organic matter content using soil type–specific depth functions. Geoderma162(1-2), 107-123.
Khormali, F., & Kehl, M. 2011. Micromorphology and development of loess-derived surface and buried soils along a precipitation gradient in northern Iran. Quaternary International, 234,109-123.
Khormali, F., Ghergherechi, S., Kehl, M., & Ayoubi, S. (2012). Soil formation in loess derived soils along a subhumid to humid climate gradient, Northeastern Iran. Geoderma, 179/180, 113-122.
Khosravi, A., K., Miran, N., Mohammadi Khajelou, Y., Khosravi Aqdam, M., Asadzadeh, F., & Mosleh, Z. (2021). Predicting the spatial distribution of soil mineral particles using OLI sensor in northwest of Iran. Environmental Monitoring and Assessment, 193, 1-13.
Lagacherie, P., & McBratney, A.B. (2006). Spatial soil information systems and spatial soil inference systems: perspectives for digital soil mapping. Developments in soil science31, 3-22.
Liaghat, M., & Khormali, F. (2011). Micromorphology of development of some loess-derived soils of
western Golestan province along a climo-topo-biosequence. J Soil Water Conserv, 18(1), 1-31. (In Persian).
Liu, F., Zhang, G., Sun, Y., Zhao, Y., Li, D. (2013). Mapping the three-dimensional distribution of soil organic matter across a subtropical hilly landscape. Soil Sci. Soc. Am. J. 77, 1241–1253.
Liu, F., Zhang, G.L., Song, X., Li, D., Zhao, Y., Yang, J., Wu, H., & Yang, F. (2020). High-resolution and three-dimensional mapping of soil texture of China. Geoderma361, 114061.
Malone, B. P., McBratney, A. B., Minasny, B., & Laslett, G.M. (2009). Mapping continuous depth functions of soil carbon storage and available water capacity. Geoderma154(1-2), 138-152.
McBratney, A. B., Mendonça Santos, M.L., & Minasny, B. (2003). On digital soil mapping. Geoderma, 117:3–52. doi:10.1016/S0016-7061(03)00223-4
Meinshausen, N., Ridgeway, G. (2006). Quantile regression forests. Journal of machine learning research. 7(6).
Minasny, B. & McBratney, A. B. (2006). A conditioned Latin hypercube method for sampling in the presence of ancillary information. Computers & geosciences32(9), 1378-1388.
Minasny, B., McBratney, A. B., Malone, B.P., Wheeler, I. (2013). Digital mapping of soil carbon. Adv. Agro, 118: 1-47.
Mitran, T., Solanky, V., Suresh, G.J., Sujatha, G., Sreenivas, K., & Ravisankar, T. (2019). Predictive mapping of surface soil texture in a semiarid region of India through geostatistical modeling. Modeling Earth Systems and Environment, 5, 645-657.
Momtaz, H.R., Jafarzadeh, A. A., Torabi, H., Oustan, S., Samadi, A., Davatgar, N., & Gilkes, R.J. (2009). An assessment of the variation in soil properties within and between landform in the Amol region, Iran. Geoderma149 (1-2), 10-18.
Moore, I.D., Gessler, P.E., Nielsen, G.A., & Peterson, G.A. (1993). Soil attribute
prediction using terrain analysis. Soil Science Society of America Journal, 57 (2): 443-452.
Mousavi, S. R., Sarmadian, F., & Rahmani, A. (2020). Modeling and Prediction of Soil Classes Using Boosting Regression Tree and Random Forests Machine Learning Algorithms in Some Part of Qazvin Plain. Iranian Journal of Soil and Water Research, 50 (10), 2525-2538. (In Persian).
Nussbaum, M., Spiess, K., Baltensweiler, A., Grob, U., Keller, A., Greiner, L., Schaepman, M.E. and Papritz, A., (2018). Evaluation of digital soil mapping approaches with large sets of environmental covariates. Soil, 4(1), 1-22.
Pahlavan-Rad, M.R., & Akbarimoghaddam, A. (2018). Spatial variability of soil texture fractions and pH in a flood plain (case study from eastern Iran). Catena160, 275-281.
Pahlavan-Rad, M.R., Toomanian, N., Khormali, F., Brungard, C.W., Bayram Komaki, C., & Bogaert, P. 2014. Updating soil survey maps using random forest and conditioned Latin hypercube sampling in the loess derived soils of northern Iran. Geoderma, 232(97): 232.
Ponce‐Hernandez, R., Marriott, F.H.C., & Beckett, P.H.T. (1986). An improved method for reconstructing a soil profile from analyses of a small number of samples. Journal of Soil Science37 (3), 455-467.
Reddy, N.N., & Das, B.S. (2023). Digital soil mapping of key secondary soil properties using pedotransfer functions and Indian legacy soil data. Geoderma429, 116265.
Renmin, Y. A. N. G., Feng, L. I. U., Zhang, G., Yuguo, Z. H. A. O., Decheng, L. I., Jinling, Y. A. N. G., Fei, Y.A. N. G., & Fan, Y. A. N. G., (2016). Mapping soil texture based on field soil moisture observations at a high temporal resolution in an oasis agricultural area. Pedosphere, 26(5), 699-708.
Rentschler, T., Gries, P., Behrens, T., Bruelheide, H., Kühn, P., Seitz, S., Shi, X., Trogisch, S., Scholten, T., & Schmidt, K. (2019). Comparison of catchment scale 3D and 2.5 D modelling of soil organic carbon stocks in Jiangxi Province, PR China. Plos one, 14(8), e0220881.
Rezaei, M., Mousavi, S.R., Rahmani, A., Zeraatpisheh, M., Rahmati, M., Pakparvar, M., Mahjenabadi, V.A.J., Seuntjens, P. & Cornelis, W., (2023). Incorporating machine learning models and remote sensing to assess the spatial distribution of saturated hydraulic conductivity in a light-textured soil. Computers and Electronics in Agriculture, 209, 107821.
Roozitalab, M.H., Toomanian, N., Ghasemi Dehkordi, V.R., & Khormali, F. (2018). Major soils, properties, and classification. The soils of Iran, 93-147.
Sahraee, N., Landi, A., & Hojati, S. (2022) Digital mapping of soil texture components in part of Khuzestan plain lands using machine learning models. Iranian Journal of Soil and Water Research, 53 (10), 2261-2276. (In Persian)
Stoorvogel, J.J., Kempen, B., Heuvelink, G.B.M., & De Bruin, S. (2009). Implementation and evaluation of existing knowledge for digital soil mapping in Senegal. Geoderma149(1-2), 161-170.
Sulaeman, Y., Minasny, B., McBratney, A.B., Sarwani, M., & Sutandi, A., 2013. Harmonizing legacy soil data for digital soil mapping in Indonesia. Geoderma, 192, 77-85.
Taghizadeh-Mehrjardi, R., Mahdianpari, M., Mohammadimanesh, F., Behrens, T., Toomanian, N., Scholten, T., & Schmidt, K. (2020). Multi-task convolutional neural networks outperformed random forest for mapping soil particle size fractions in central Iran. Geoderma376, 114552.
Taghizadeh-Mehrjardi, R., Minasny, B., Sarmadian, F., & Malone, B. P. (2014). Digital mapping of soil salinity in Ardakan region, central Iran. Geoderma, 213, 15-28.
Taghizadeh-Mehrjardi, R., Nabiollahi, K., Kerry, R. (2016). Digital mapping of soil organic carbon at multiple depths using different data mining techniques in Baneh region, Iran. Geoderma, 266: 98-110.
Thompson, J. A., Roecker, S., Grunwald, S., & Owens, P. R. (2012). Digital soil mapping: Interactions with and applications for hydropedology. In: H. Lin (ed). Hydropedology, 665-709. Amsterdam: Academic Press.
Vaysse, K., Lagacherie, Ph. (2017). Using quantile regression forest to estimate uncertainty of digital soil mapping products. Geoderma, 291: 55–64.
Wilding, L.P. (1985). Spatial variability: its documentation, accomodation and implication to soil surveys. In Soil spatial variability, Las Vegas NV, 30 November-1 December 1984, 166-194.
Xiao, J., Shen, Y., Tateishi, R., & Bayaer, W. (2006). Development of topsoil grain size index for monitoring desertification in arid land using remote sensing. International Journal of Remote Sensing27 (12), 2411-2422.
Zeraatpisheh, M., Ayoubi, S., Jafari, A., Tajik, S., & Finke, P. (2019). Digital mapping of soil properties using multiple machine learning in a semi-arid region, central Iran. Geoderma338, 445-452.
Zeraatpisheh, M., Jafari, A., Bodaghabadi, M.B., Ayoubi, S., Taghizadeh-Mehrjardi, R., Toomanian, N., Kerry, R., & Xu, M., (2020). Conventional and digital soil mapping in Iran: Past, present, and future. Catena, 188, 104424.