Mapping soil organic carbon stocks of different land use types in the Southern Moscow region by applying machine learning to legacy data

Cover Page

Cite item

Full Text

Abstract

This study presents the result of topsoil (0—10 cm) soil organic carbon (SOC) mapping in two areas of Moscow Region (2007 status): 1096 km2 — Podolsky District, and 1101 km2 — Serpukhovsky District. Based on 2007 legacy soil sampling data (n = 282) within these areas, we have created a statistical model between the target variable (SOC stocks, kg/m2) and numerous covariates (legacy maps and remote sensing data). GBM model has explained 56% of soil organic carbon stocks variability. Differences in stocks within different land use types were shown quantitatively. At the same time, the spectral reflectance in the near infrared band (B5) of Landsat‑5 TM made the greatest contribution in explaining the differences within individual types (among fallow lands and urbanized areas), and the spectral index NDVI has explained the spatial variability of soil organic carbon among forest ecosystems. The root mean square error of cross-­validation (RMSEcv = 0.67 kg/m2) was chosen to describe the uncertainty of soil organic carbon stock prediction. According to the model, the total soil organic carbon stocks in the upper 10 cm soil layer of the Podolsky District were 2.65 ± 0.72 Tg, for the Serpukhovsky District — 2.77 ± 0.73 Tg.

Full Text

Table 1
List of covariates used to predict SOC in the 0—10 cm of soils

 SCOPRAN

 Predictor (code)

 Source

 Link

 R

 elevation (elev)

 DEM SRTM 30m

 

 R

 slope (slope)

 —

 

 R

 aspect (aspect)

 —

 

 R, N

 Topographic Wetness Index (twi)

 —

 [38]

 R

 flow direction (flowdir)

 —

 [39]

 R

 flow accumulation (flowacc)

 —

 [39]

 P, A, N

 B1_BLUE (b1)

 Landsat‑5 TM (median)

 

 P, A, N

 B2_GREEN (b2)

 —

 

 P, A, N

 B3_RED (b3)

 —

 

 P, A, N

 B4_NIR1 (b4)

 —

 

 P, A, N

 B5_NIR2 (b5)

 —

 

 P, A, N

 B7_SWIR (b7)

 —

 

 P, A, N

 B3/B2 (b2b3_ratio)

 —

 [40]

 P, A, N

 B5/B4 (b5b4_ratio)

 —

 [40]

 O

 NDVI (ndvi)

 —

 [41]

 O

 SAVI (savi)

 —

 

 C

 MNDWI (mndwi)

 —

 [42]

 P, N

 NDBI (ndbi)

 —

 [43]

 O, A, S

 land use map (landcover)

 —

 

 S

 soil type (soils)

 Soil map of the Moscow region

 [32]

 P, S

 grain-size composition (gms)

 Soil map of the Moscow region

 [32]

Note. R — relief; N — space; A — age; P — parent material; O — organisms; C — climate; S — soil.
Source: compiled by Yu.A. Dvornikov.

Fig. 1. Distribution of topsoil SOC stocks of different land use types according to legacy soil survey data: UT — urbanized territories
Source: compiled by Yu.A. Dvornikov.

Fig. 2. Relative importance of predictors, %, explaining the variability of SOC in the upper 10cm layer (a), and comparison of measured and predicted SOC stocks values (0—10 cm) in two districts of the Moscow region (b). UT — urbanized territories
Source: compiled by Yu.A. Dvornikov.

Fig. 3. Spatial distribution of SOC stocks in the topsoil (0—10 cm) for Podolskiy and Serpukhovskiy districts of Moscow region (predicted values as of 2007)
Source: compiled by Yu.A. Dvornikov.

Table 2
Total SOC stocks in the upper 10 cm soil layer of Podolskiy and Serpukhovskiy administrative districts (as of 2007) for four main land use types

 District

 Land use

 Predicted SOC stocks, Tg

 Limit of uncertainty, Tg

 Lower

 Upper

 Podolskiy

 Forest ecosystems

 1.48

 1.11

 1.84

 Fallow lands

 0.62

 0.43

 0.81

Urbanized territories

 0.09

 0.07

 0.12

 Croplands

 0.46

 0.32

 0.6

 Serpukhovskiy

 Forest ecosystems

 1.68

 1.29

 2.07

 Fallow lands

 0.84

 0.58

 1.1

Urbanized territories

 0.05

 0.04

 0.07

 Croplands

 0.2

 0.13

 0.26

Source: compiled by Yu.A. Dvornikov.

×

About the authors

Yury A. Dvornikov

RUDN University; Institute of Physicochemical and Biological Problems of Soil Science RAS

Author for correspondence.
Email: ydvornikow@gmail.com
ORCID iD: 0000-0003-3491-4487
SPIN-code: 8020-3292

Candidate of Geological and Mineralogical Science, Associate Professor, Laboratory of Smart Urban Nature, Аgroengineering department, Agrarian and Technological Institute, RUDN University; Researcher, Laboratory of Carbon Monitoring in Terrestrial Ecosystems, Institute of Physicochemical and Biological Problems in Soil Science RAS

8/2 Miklukho-Maklaya st., Moscow, 117198, Russian Federation; 2 Institutskaya st., Pushchino, Moscow region, 142290, Russian Federation

Lukyan A. Mirniy

Institute of Physicochemical and Biological Problems of Soil Science RAS

Email: mirluk@yandex.ru
Engineer, laboratory of Carbon Monitoring in Terrestrial Ecosystems 2 Institutskaya st., Pushchino, Moscow region, 142290, Russian Federation

Ekaterina S. Mukvich

Institute of Physicochemical and Biological Problems of Soil Science RAS

Email: katerinamykvitc@mail.ru
ORCID iD: 0009-0004-5378-8775
SPIN-code: 2640-0913

PhD student, Junior Researcher, Laboratory of Carbon Monitoring in Terrestrial Ecosystems

2 Institutskaya st., Pushchino, Moscow region, 142290, Russian Federation

Kristina V. Ivashchenko

Institute of Physicochemical and Biological Problems of Soil Science RAS

Email: ivashchenko.kv@gmail.com
ORCID iD: 0000-0001-8397-158X
SPIN-code: 1388-1561

Candidate of Sciences in Biology, Senior Researcher, Laboratory of Carbon Monitoring in Terrestrial Ecosystems

2 Institutskaya st., Pushchino, Moscow region, 142290, Russian Federation

References

  1. Batjes NH. Total carbon and nitrogen in the soils of the world. European Journal of Soil Science. 1996;47(2):151—63. doi: 10.1111/j.1365-2389.1996.tb01386.x
  2. Chernova OV, Golozubov OM, Alyabina IO, Schepaschenko DG. Integrated approach to spatial assessment of soil organic carbon in the Russian Federation. Eurasian Soil Science. 2021;54(3):325—336. doi: 10.1134/s1064229321030042
  3. Minasny B, Malone BP, McBratney AB, Field DJ, Odeh I, Padarian J, et al. Soil carbon 4 per mille. Geoderma. 2017;292:59—86. doi: 10.1016/j.geoderma.2017.01.002
  4. de Gruijter JJ, Brus DJ, Bierkens MFP, Knotters M. Sampling for natural resource monitoring. Berlin (Germany) etc.: Springer; 2006.doi: 10.1007/3-540-33161-1
  5. Kudeyarov VN, Zavarzin GA, Blagodatskij SA, Borisov AV, Voronin PY, Demkin VA, et al. Puly i potoki ugleroda v nazemnyh ekosistemah Rossii [Carbon pools and flows in terrestrial ecosystems of Russia]. Moscow: Nauka publ.; 2007. (In Russ).
  6. Zhang Z, Xia L, Zhao Z, Zhao F, Hou G, Wu S, et al. How land use transitions contribute to the soil organic carbon accumulation from 1990 to 2020. Remote Sensing. 2024;16(7):1308. doi: 10.3390/rs16071308
  7. Kurganova IN, Lopes de Gerenju VO, Mjakshina TN, Sapronov DV, Savin IJu, Shorohova EV. Carbon balance in forest ecosystems of Southern Moscow region under rising aridity of climate. Lesovedenie. 2016;(5):332—345. (In Russ).
  8. McBratney AB, Mendonça Santos ML, Minasny B. On digital soil mapping. Geoderma. 2003;117(1—2):3—52. doi: 10.1016/s0016-7061 (03) 00223-4
  9. Minasny B, McBratney AB, Malone BP, Wheeler I. Digital Mapping of Soil Carbon. Advances in Agronomy. 2013;118:1—47. doi: 10.1016/b978-0-12-405942-9.00001-3
  10. Savin IY, Zhogolev AV, Prudnikova EY. Modern Trends and Problems of Soil Mapping. Eurasian Soil Science. 2019;52(5):471—480. doi: 10.1134/s1064229319050107
  11. Wadoux AMJC, Minasny B, McBratney AB. Machine learning for digital soil mapping: Applications, challenges and suggested solutions. Earth-­Science Reviews. 2020;210:103359. doi: 10.1016/j.earscirev.2020.103359
  12. Li J, Heap AD. Spatial interpolation methods applied in the environmental sciences: A review. Environmental Modelling & Software. 2014;53:173—189. doi: 10.1016/j.envsoft.2013.12.008
  13. Gopp NV, Meshalkina JL, Narykova AN, Plotnikova AS, Chernova OV. Mapping of soil organic carbon content and stock at the regional and local levels: the analysis of modern methodological approaches. Forest science issues. 2023;6(1):1—59. (In Russ.). doi: 10.31509/2658-607x‑202361-120
  14. Veronesi F, Schillaci C. Comparison between geostatistical and machine learning models as predictors of topsoil organic carbon with a focus on local uncertainty estimation. Ecological Indicators. 2019;101:1032—1044. doi: 10.1016/j.ecolind.2019.02.026
  15. Florinsky IV. The Dokuchaev hypothesis as a basis for predictive digital soil mapping (on the 125th anniversary of its publication). Eurasian Soil Science. 2012;45(4):445—451. (In Russ.). doi: 10.1134/S1064229312040047
  16. Ivashchenko KV, Sushko SV, Dvornikov YA, Mirny LA, Orlova LV, Ananyeva ND, et al. Soil Organic Carbon Stocks under No-­Tillage in the Middle Volga Region. Agrohimia. 2023;(12):47—56. (In Russ.). doi: 10.31857/s0002188123110066
  17. Dvornikov Y, Slukovskaya M, Yaroslavtsev A, Meshalkina J, Ryazanov A, Sarzhanov D, et al. High-resolution mapping of soil pollution by Cu and Ni at a polar industrial barren area using proximal and remote sensing. Land Degradation & Development. 2022;33(10):1731—1744. doi: 10.1002/ldr.4261
  18. Florinsky IV, Eilers RG, Manning GR, Fuller LG. Prediction of soil properties by digital terrain modelling. Environmental Modelling and Software. 2002;17(3):295—311. doi: 10.1016/s1364-8152 (01) 00067-6
  19. Dvornikov YA, Vasenev VI, Romzaykina ON, Grigorieva VE, Litvinov YA, Gorbov SN, et al. Projecting the urbanization effect on soil organic carbon stocks in polar and steppe areas of European Russia by remote sensing. Geoderma. 2021;399:115039. doi: 10.1016/j.geoderma.2021.115039
  20. Vasenev VI, Stoorvogel JJ, Vasenev II, Valentini R. How to map soil organic carbon stocks in highly urbanized regions? Geoderma. 2014;226–227:103—115. doi: 10.1016/j.geoderma.2014.03.007
  21. Dobrovolsky GV, Urusevskaya IS. Geografiya pochv [Soil geography]. Moscow: Nauka publ.; 2006. (In Russ.).
  22. Poggio L, de Sousa LM, Batjes NH, Heuvelink GBM, Kempen B, Ribeiro E, et al. SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty. SOIL. 2021;7(1):217—240. doi: 10.5194/soil‑7-217-2021
  23. Chinilin AV, Savin IY. Estimation of organic carbon content in Russian soils using ensemble machine learning. Vestnik Moskovskogo universiteta. Seriya 5, Geografiya. 2022;(6):49—63. (In Russ.). doi: 10.55959/MSU0579-9414-5-2022-6-49-63
  24. Suleymanov A, Abakumov E, Nizamutdinov T, Polyakov V, Shevchenko E, Makarova M. Soil organic carbon stock retrieval from Sentinel‑2A using a hybrid approach. Environmental Monitoring and Assessment. 2024;196(1):23. doi: 10.1007/s10661-023-12172‑y
  25. Lado LR, Hengl T, Reuter HI. Heavy metals in European soils: A geostatistical analysis of the FOREGS Geochemical database. Geoderma. 2008;148(2):189—199. doi: 10.1016/j.geoderma.2008.09.020
  26. Wadoux AMJC, Brus DJ. How to compare sampling designs for mapping? European Journal of Soil Science. 2020;72(1):35—46. doi: 10.1111/ejss.12962
  27. Wadoux AMJC, Brus DJ, Heuvelink GBM. Sampling design optimization for soil mapping with random forest. Geoderma. 2019;355:113913. doi: 10.1016/j.geoderma.2019.113913
  28. Hengl T. Finding the right pixel size. Computers & Geosciences. 2006;32(9):1283—1298. doi: 10.1016/j.cageo.2005.11.008
  29. Romanenkov VA, Meshalkina JL, Gorbacheva AY, Krenke AN, Petrov IK, Golozubov OM, et al. Maps of Soil Organic Carbon Sequestration Potential in the Russian Croplands. Eurasian Soil Science. 2024;57(5):737—750. doi: 10.1134/s106422932360375x
  30. Beck HE, Zimmermann NE, McVicar TR, Vergopolan N, Berg A, Wood EF. Present and future Köppen-­Geiger climate classification maps at 1‑km resolution. Scientific Data. 2018;5(1): 180214. doi: 10.1038/sdata.2018.214
  31. Boldyreva VE, Golozubov OM, Litvinov YA, Minaeva EN, Pulin AV. Tsifrovaya srednemasshtabnaya Pochvennaya karta Moskovskogo regiona [Digital meso-scale soil map of Moscow Oblast’]. Mapping division of IS SSDB. Available from: https://soil-db.ru/map?name=moscow-­region [Accessed 15th April 2024] (In Russ.).
  32. Gavrilenko EG. Biologicheskiye svoystva pochvy dlya ikh ekologo-­ekonomicheskoi otsenki (na primere Serpukhovskogo i Podol’skogo raionov Moskovskoi oblasti) [Biological properties of soil for their ecological and economic assessment (using the example of Serpukhov and Podolsk districts of the Moscow region)] [Dissertation] Moscow; 2013. (In Russ.).
  33. Prévost M. Predicting Soil Properties from Organic Matter Content following Mechanical Site Preparation of Forest Soils. Soil Science Society of America Journal. 2004;68(3):943—949. doi: 10.2136/sssaj2004.9430
  34. Breiman L. Random Forests. Machine Learning. 2001;45:5—32. doi: 10.1023/A:1010933404324
  35. Gorelick N, Hancher M, Dixon M, Ilyushchenko S, Thau D, Moore R. Google Earth Engine: Planetary-­scale geospatial analysis for everyone. Remote Sensing of Environment. 2017;202:18—27. doi: 10.1016/j.rse.2017.06.031
  36. Beven KJ, Kirkby MJ. A physically based, variable contributing area model of basin hydrology. Hydrological Sciences Bulletin. 1979;24(1):43—69. doi: 10.1080/02626667909491834
  37. Florinsky IV. An illustrated introduction to geomorphometry. Electronic scientific Edition Almanac Space and Time. 2016;11(1):18. (In Russ.).
  38. Meddens AJH, Hicke JA, Vierling LA, Hudak AT. Evaluating methods to detect bark beetle-­caused tree mortality using single-date and multi-date Landsat imagery. Remote Sensing of Environment. 2013;132:49—58. doi: 10.1016/j.rse.2013.01.002
  39. Tucker CJ. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing of Environment. 1979;8(2):127—150. doi: 10.1016/0034-4257 (79) 90013-0
  40. Xu H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. International Journal of Remote Sensing. 2006;27(14):3025—3033. doi: 10.1080/01431160600589179
  41. Zha Y, Gao J, Ni S. Use of normalized difference built-up index in automatically mapping urban areas from TM imagery. International Journal of Remote Sensing. 2003;24(3):583—594. doi: 10.1080/01431160304987

Supplementary files

Supplementary Files
Action
1. Fig. 1. Distribution of topsoil SOC stocks of different land use types according to legacy soil survey data: UT — urbanized territories
Source: compiled by Yu.A. Dvornikov.

Download (53KB)
Indexing metadata
2. Fig. 2. Relative importance of predictors, %, explaining the variability of SOC in the upper 10cm layer (a), and comparison of measured and predicted SOC stocks values (0—10 cm) in two districts of the Moscow region (b). UT — urbanized territories
Source: compiled by Yu.A. Dvornikov.

Download (85KB)
Indexing metadata
3. Fig. 3. Spatial distribution of SOC stocks in the topsoil (0—10 cm) for Podolskiy and Serpukhovskiy districts of Moscow region (predicted values as of 2007)
Source: compiled by Yu.A. Dvornikov.

Download (314KB)
Indexing metadata

Copyright (c) 2024 Dvornikov Y.A., Mirniy L.A., Mukvich E.S., Ivashchenko K.V.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies