High Accuracy Geochemical Map Generation Method by a Spatial Autocorrelation-Based Mixture Interpolation Using Remote Sensing Data
Abstract
:1. Introduction
1.1. Background
1.2. Why Spatial Information Can be Coupled into an Ml Model Only by Data Augmentation
1.2.1. Spatial Information Calculated from Kriging Interpolation
1.2.2. Merge Spatial Information into RF Model
2. Materials and Methods
2.1. Study Area and Target Variable
2.2. Predictor Variables–High-Resolution Remote Sensing Data
2.3. Spatial Autocorrelation-Based Mixture Interpolation Algorithm
2.3.1. The Process of the SABAMIN Algorithm
2.3.2. How to Generate Pseudo Training Data in Our Case
2.4. Algorithm Validity Test
3. Results
3.1. Cross-Validation Test of Measured Data
3.2. Blind Test in T Area
3.3. Temporal Stability Test in P Areas
3.4. The Predictor Importance Test
4. Discussion
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Pal, M.; Rasmussen, T.; Porwal, A. optimized lithological mapping from multispectral and hyperspectral remote sensing images using fused multi-classifiers. Remote Sens. 2020, 12, 177. [Google Scholar] [CrossRef] [Green Version]
- Kirkwood, C.; Cave, M.; Beamish, D.; Grebby, S.; Ferreira, A. A machine learning approach to geochemical mapping. J. Geochem. Explor. 2016, 167, 49–61. [Google Scholar] [CrossRef] [Green Version]
- Gahegan, M. On the application of inductive machine learning tools to geographical analysis. Geogr. Anal. 2000, 32, 113–139. [Google Scholar] [CrossRef]
- Mather, P.M.; Koch, M. Computer Processing of Remotely-Sensed Images: An introduction; John Wiley & Sons: Hoboken, NJ, USA, 2011; pp. 1–75. [Google Scholar]
- Cracknell, M.J.; Reading, A.M. Geological mapping using remote sensing data: A comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information. Comput. Geosci. 2014, 63, 22–33. [Google Scholar] [CrossRef] [Green Version]
- Sergeev, A.P.; Buevich, A.G.; Baglaeva, E.M.; Shichkin, A.V. Combining spatial autocorrelation with machine learning increases prediction accuracy of soil heavy metals. Catena 2019, 174, 425–435. [Google Scholar] [CrossRef]
- Tanner, M.A.; Wong, W.H. The calculation of posterior distributions by data augmentation. J. Am. Stat. Assoc. 1987, 82, 528–540. [Google Scholar] [CrossRef]
- Liu, J.S.; Wu, Y.N. Parameter expansion for data augmentation. J. Am. Stat. Assoc. 1999, 94, 1264–1274. [Google Scholar] [CrossRef]
- Perez, L.; Wang, J. The effectiveness of data augmentation in image classification using deep learning. arXiv 2017, arXiv:1712.04621. Available online: https://arxiv.org/abs/1712.04621v1 (accessed on 7 December 2016).
- Kleijnen, J.P. Kriging metamodeling in simulation: A review. Eur. J. Oper. Res. 2009, 192, 707–716. [Google Scholar] [CrossRef] [Green Version]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Huang, C.; Shibuya, A. Approach for generating high accuracy machine learning model for high resolution geochemical map completion using remote sensing data: Case study of Arizona, USA. In Earth Resources and Environmental Remote Sensing/GIS Applications X; SPIE: Strasbourg, France, 2019; p. 111560F. [Google Scholar] [CrossRef]
- Shepard, D. A two-dimensional interpolation function for irregularly-spaced data. In Proceedings of the 1968 23rd ACM National Conference, New York, NY, USA, 27–29 August 1968; ACM: New York, NY, USA, 1968; pp. 517–524. [Google Scholar]
- McBratney, A.B.; Webster, R. Choosing functions for semi-variograms of soil properties and fitting them to sampling estimates. J. Soil Sci. 1986, 37, 617–639. [Google Scholar] [CrossRef]
- Cressie, N. Statistics for Spatial Data; Revised Edition; John Wiley & Sons: Hoboken, NJ, USA, 1993; pp. 1–275. [Google Scholar]
- Lang, J.R.; Titley, S.R. Isotopic and geochemical characteristics of laramide magmatic systems in Arizona and implications for the genesis of porphyry copper deposits. Econ. Geol. 1998, 93, 138–170. [Google Scholar] [CrossRef]
- Bodnar, R.J.; Beane, R.E. Temporal and spatial variations in hydrothermal fluid characteristics during vein filling in preore cover overlying deeply buried porphyry copper-type mineralization at Red Mountain, Arizona. Econ. Geol. 1980, 75, 876–893. [Google Scholar] [CrossRef]
- Manske, S.L.; Paul, A.H. Geology of a major new porphyry copper center in the Superior (Pioneer) district, Arizona. Econ. Geol. 2002, 97, 197–220. [Google Scholar] [CrossRef]
- Abrams, M.J.; Brown, D.; Lepley, L.; Sadowski, R. Remote sensing for porphyry copper deposits in southern Arizona. Econ. Geol. 1983, 78, 591–604. [Google Scholar] [CrossRef]
- Smith, S.M. National Geochemical Database Reformatted Data from the National Uranium Resource Evaluation (NURE) Hydrogeochemical and Stream Sediment Reconnaissance (HSSR) Program; US Geological Survey, US Dept. of the Interior: Washington, DC, USA, 1997; pp. 97–492. Available online: https://pubs.usgs.gov/of/1997/ofr-97-0492/ (accessed on 7 December 2016).
- Smith, S.M. A Manual for Interpreting USGS-Reformatted NURE HSSR Data Files; USGS Publications Warehouse: Reston, VA, USA, March 2006. Available online: https://pubs.usgs.gov/of/1997/ofr-97-0492/nure_man.htm (accessed on 7 December 2016).
- Schruben, P.G.; Arndt, R.E.; Bawiec, W.J. Geology of the Conterminous United States at 1: 2,500,000 Scale a Digital Representation of the 1974 PB King and HM Beikman Map; Technical Report No. 11; The Survey; For sale by USGS Map Distribution; USGS: Reston, VA, USA, 1998. Available online: https://pubs.usgs.gov/dds/dds11/ (accessed on 7 December 2016).
- Mason, G.T.; Arndt, R.E. Mineral Resources Data System (MRDS); No. 20; USGS Numbered Series; USGS: Reston, VA, USA, 1996. [CrossRef]
- McFaul, E.J.; Mason, G.T.; Ferguson, W.B.; Lipin, B.R. U.S. Geological Survey Mineral Databases; Technical Report No. 50; MRDS and MAS/MILS: Reston, VA, USA, 2000. [CrossRef]
- Bankey, V.; Cuevas, A.; Daniels, D.; Finn, C.A.; Israel, H.; Hill, P.; Kucks, R.; Miles, W.; Pilkington, M.; Roberts, C.; et al. Digital Data Grids for the Magnetic Anomaly Map of North America; USGS Open-File Report 02-414; USGS: Reston, VA, USA, 2002. Available online: https://pubs.usgs.gov/of/2002/ofr-02-414 (accessed on 25 October 2002).
- Van der Meer, F.D.; Van der Werff, H.M.; Van Ruitenbeek, F.J.; Hecker, C.A.; Bakker, W.H.; Noomen, M.F.; Van der Meijde, M.; Carranza, E.J.M.; Van de Smeth, J.B.; Woldai, T. Multi-and hyperspectral geologic remote sensing: A review. Int. J. Appl. Earth Obs. 2012, 14, 112–128. [Google Scholar] [CrossRef]
- Ninomiya, Y.; Fu, B. Spectral indices for lithologic mapping with ASTER thermal infrared data applying to a Part of Beishan Mountains. In Proceedings of the IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No. 01CH37217), IEEE, Sidney, Australia, 9–13 July 2001; pp. 2988–2990. [Google Scholar]
- Lindsay, M.D.; Betts, P.G.; Ailleres, L. Data fusion and porphyry copper prospectivity models, southeastern Arizona. Ore Geol. Rev. 2014, 61, 120–140. [Google Scholar] [CrossRef]
- Yigit, O. A prospective sector in the tethyan metallogenic belt: Geology and geochronology of mineral deposits in the biga peninsula, NW Turkey. Ore Geol. Rev. 2012, 46, 118–148. [Google Scholar] [CrossRef]
- Abedi, M.; Norouzi, G.H. Integration of various geophysical data with geological and geochemical data to determine additional drilling for copper exploration. J. Appl. Geophys. 2012, 83, 35–45. [Google Scholar] [CrossRef]
- Edelsbrunner, H.; Kirkpatrick, D.; Seidel, R. On the shape of a set of points in the plane. IEEE Trans. Inf. Theory 1983, 29, 551–559. [Google Scholar] [CrossRef] [Green Version]
- Guibas, L.; Stolfi, J. Primitives for the manipulation of general subdivisions and the computation of Voronoi. ACM Trans. Graph. 1985, 4, 74–123. [Google Scholar] [CrossRef]
- Sutherland, I.E.; Sproull, R.F.; Schumacker, R.A. A characterization of ten hidden-surface algorithms. ACM Comput. Surv. 1974, 6, 1–55. [Google Scholar] [CrossRef]
- Nilufar, S.; Morrow, A.A.; Lee, J.M.; Perkins, T.J. FiloDetect: Automatic detection of filopodia from fluorescence microscopy images. BMC Syst. Biol. 2013, 7, 66. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Ayuso, R.A.; Barton, M.D.; Blakely, R.J.; Bodnar, R.J.; Dilles, J.H.; Gray, F.; Graybeal, F.T.; Mars, J.C.; McPhee, D.K.; Seal, R.R.; et al. Porphyry Copper Deposit Model: Chapter B in Mineral Deposit Models for Resource Assessment; Technical Report No. 2010-5070-B; US Geological Survey: Reston, VA, USA, 2010; pp. 1–130. Available online: https://pubs.usgs.gov/sir/2010/5070/b/ (accessed on 7 December 2016).
- Tran, T.; Pham, T.; Carneiro, G.; Palmer, L.; Reid, I. A bayesian data augmentation approach for learning deep models. In Advances in Neural Information Processing Systems; NIPS: Long Beach, CA, USA, December 2017; pp. 2797–2806. [Google Scholar]
- Salamon, J.; Bello, J.P. Deep convolutional neural networks and data augmentation for environmental sound classification. IEEE Signal Process. Lett. 2017, 24, 279–283. [Google Scholar] [CrossRef]
- Kim, H.; Willers, J.L.; Kim, S. The curvature interpolation method for surface reconstruction for geospatial point cloud data. Int. J. Remote Sens. Appl. 2020, 41, 1512–1541. [Google Scholar] [CrossRef]
- Chen, Y.; Wu, W. Mapping mineral prospectivity using an extreme learning machine regression. Ore Geol. Rev. 2017, 80, 200–213. [Google Scholar] [CrossRef]
- Carranza, E.J.M.; Laborte, A.G. Random forest predictive modeling of mineral prospectivity with small number of prospects and data with missing values in Abra (Philippines). Comput. Geosci. 2015, 74, 60–70. [Google Scholar] [CrossRef]
- Abedi, M.; Norouzi, G.H.; Bahroudi, A. Support vector machine for multi-classification of mineral prospectivity areas. Comput. Geosci. 2012, 46, 272–283. [Google Scholar] [CrossRef]
- Lloyd, C.D. Local Models for Spatial Analysis; CRC Press: Boca Raton, FL, USA, 2011; p. 336. [Google Scholar]
No. | G/L Type | No. | G/L Type | No. | G/L Type | No. | G/L Type |
---|---|---|---|---|---|---|---|
1 | Water | 14 | Eocene | 27 | Upper Mesozoic eugeosynclinal | 40 | Upper Paleozoic eugeosynclinal |
2 | Quaternary | 15 | Paleocene continental | 28 | Jurassic | 41 | Upper Paleozoic clastic wedge facies |
3 | Quaternary volcanic rocks | 16 | Navarro Group | 29 | Lower Mesozoic volcanic rocks | 42 | Lower Paleozoic |
4 | Pliocene continental | 17 | Taylor Group | 30 | Jurassic granitic rocks | 43 | Cambrian |
5 | Pliocene volcanic rocks | 18 | Latest Cretaceous granitic | 31 | Lower Jurassic and upper Triassic | 44 | Z sedimentary rocks |
6 | Pliocene felsic volcanic rocks | 19 | Austin and Eagle Ford Groups | 32 | Lower Mesozoic | 45 | Y sedimentary rocks |
7 | Miocene continental | 20 | Upper Cretaceous | 33 | Lower Mesozoic eugeosynclinal | 46 | Younger Y granitic rocks |
8 | Tertiary intrusive rocks | 21 | Cretaceous continental | 34 | Triassic | 47 | Older Y granitic rocks |
9 | Oligocene continental | 22 | Cretaceous volcanic rocks | 35 | Permian | 48 | X metasedimentary rocks |
10 | Miocene volcanic rocks | 23 | Cretaceous granitic rocks | 36 | Upper part of Leonardian Series | 49 | X granitic rocks |
11 | Miocene felsic volcanic rocks | 24 | Woodbine and Tuscaloosa groups | 37 | Lower part of Leonardian Series | 50 | Orthogneiss and paragneiss |
12 | Eocene continental | 25 | Fredericksburg Group | 38 | Wolfcampian Series continental | ||
13 | Lower Tertiary volcanic rocks | 26 | Lower Cretaceous | 39 | Upper Paleozoic |
Area | Range | Resolution |
---|---|---|
Ta | (W111.7200, N33.5638)–(W110.9440, N32.7958) | 110 × 110 |
Tb | (W111.7840, N32.5558)–(W110.9120, N31.6838) | 110 × 110 |
Tc | (W110.7840, N32.7158)–(W109.9920, N31.9238) | 110 × 110 |
Pa | (W114.3840, N35.4758)–(W113.9840, N35.0758) | 50 × 50 |
Pb | (W109.6240, N33.2758)–(W109.2240, N32.8758) | 50 × 50 |
Pc | (W112.2640, N34.5558)–(W111.8640, N34.1558) | 50 × 50 |
Group | No. | Predictor | Predictor Description |
---|---|---|---|
Geo-magnetic | 1 | Mag_AS | Analytic signal processed geomagnetic data |
2 | Mag_RTP | Reduction to pole processed geomagnetic data | |
3 | Mag_TMI | The residual of international geomagnetic reference field (IGRF) | |
4 | Mag_VD | Vertical first derivative processed geomagnetic data | |
DEM | 5 | Altitude | The altitude of the Earth surface, from 0–2500 m |
6 | Slope | The elevation of the Earth surface | |
ASTER Band | 7 | Band_1 | The ASTER band 1 sensor data |
8 | Band_2 | The ASTER band 2 sensor data | |
9 | Band_3N | The ASTER band 3N sensor data | |
10 | Band_4 | The ASTER band 4 sensor data | |
11 | Band_5 | The ASTER band 5 sensor data | |
12 | Band_6 | The ASTER band 6 sensor data | |
13 | Band_7 | The ASTER band 7 sensor data | |
14 | Band_8 | The ASTER band 8 sensor data | |
15 | Band_9 | The ASTER band 9 sensor data | |
16 | Band_10 | The ASTER band 10 sensor data | |
17 | Band_11 | The ASTER band 11 sensor data | |
18 | Band_12 | The ASTER band 12 sensor data | |
19 | Band_13 | The ASTER band 13 sensor data | |
20 | Band_14 | The ASTER band 14 sensor data | |
21 | Band_1_R | The reverse of 7 | |
22 | Band_2_R | The reverse of 8 | |
23 | Band_3_R | The reverse of 9 | |
24 | Band_4_R | The reverse of 10 | |
25 | Band_5_R | The reverse of 11 | |
26 | Band_6_R | The reverse of 12 | |
27 | Band_7_R | The reverse of 13 | |
28 | Band_8_R | The reverse of 14 | |
29 | Band_9_R | The reverse of 15 | |
30 | Band_10_R | The reverse of 16 | |
31 | Band_11_R | The reverse of 17 | |
32 | Band_12_R | The reverse of 18 | |
33 | Band_13_R | The reverse of 19 | |
34 | Band_14_R | The reverse of 20 | |
ASTER lithological index | 35 | R_Ferric_iron | Feature index of Fe3+, =8/7 |
36 | R_Ferrous_iron | Feature index of Fe2+, =11/9 + 7/8 | |
37 | R_Laterite | Feature index of laterite, =10/11 | |
38 | R_Gosan | Feature index of gosan, =10/8 | |
39 | R_Ferrous_Silica | Feature index of ferrous silicates, mainly Fe oxide Cu-Au alteration, =11/10 | |
40 | R_Ferric_oxides | Feature index of ferric oxides, =10/9 | |
41 | R_Carbonate | Feature index of carbonate/chlorite/epidote, =(13 + 15)/14 | |
42 | R_Epidote | Feature index of epidote/chlorite/amphibole, =(12 + 15)/(13 + 14) | |
43 | R_MgOH | Feature index of Amphibole/MgOH, =(12 + 15)/14 | |
44 | R_Amphibole | Feature index of amphibole, =12/14 | |
45 | R_Carbonate2 | Feature index of carbonate, =19/20 | |
46 | R_Dolomite | Feature index of dolomite, =(12 + 14)/13 | |
47 | R_Sericite | Feature index of sericite/muscovite/illite/smectite, =(11 + 13)/12 | |
48 | R_Alunite | Feature index of alunite/kaolinite/pyrophyllite, =(10 + 1 2)/11 | |
49 | R_phengitic | Feature index of phengitic, =11/12 | |
50 | R_Muscovite | Feature index of muscovite, =13/12 | |
51 | R_Kaolinite | Feature index of kaolinite, =13/11 | |
52 | R_Quartz | Feature index of quartz rich rocks, =20/18 | |
53 | R_Basic_deg | Feature index of basic degree index of SiO2, =18/19 | |
54 | R_SiO2 | Feature index of SiO2, =19/18 | |
55 | R_Siliceous_rock | Feature index of siliceous rocks, =172/(16 × 18) | |
56 | R_Silica1 | Feature index of the first pattern of silica, =17/16 | |
57 | R_Silica2 | Feature index of the second pattern of silica, =17/18 | |
58 | R_Silica3 | Feature index of the third pattern of silica, =19/16 | |
59 | R_Vegetation | Feature index of vegetation, =9/8 | |
60 | R_Clay | Feature index of clay, =(11 × 13)/122 | |
61 | R_NDVI | Feature index of NDVI, =(9 − 8)/(9 + 8) | |
Coordinates | 62 | x | Longitude |
63 | y | Latitude |
Area | RMSE | p-Value | Rt | p-Value | ||
---|---|---|---|---|---|---|
RF | SABAMIN | RF | SABAMIN | |||
Ta | 172.5 ± 5.5 | 133.9 ± 3.9 | 0.000 ** | 0.546 ± 0.052 | 0.629 ± 0.050 | 0.000 ** |
Tb | 419.0 ± 10.5 | 410.6 ± 9.0 | 0.001 * | 0.757 ± 0.032 | 0.779 ± 0.030 | 0.005 * |
Tc | 431.2 ± 8.8 | 398.6 ± 7.3 | 0.000 ** | 0.737 ± 0.038 | 0.756 ± 0.020 | 0.021 * |
Area | Rp | ||
---|---|---|---|
tem1 vs. tem2 | tem1 vs. tem3 | tem2 vs. tem3 | |
Pa | 0.681 ** | 0.347 * | 0.548 * |
Pb | 0.756 ** | 0.710 ** | 0.834 ** |
Pc | 0.743 ** | 0.581 ** | 0.562 ** |
Group | Predictor Number | ImG (Mean ± SD) |
---|---|---|
Geomagnetic | 4 | 0.088 ± 0.012 |
DEM | 2 | 0.088 ± 0.018 |
ASTER Band | 28 | 0.210 ± 0.023 |
ASTER Lithological index | 27 | 0.457 ± 0.025 |
Coordinates | 2 | 0.158 ± 0.021 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Huang, C.; Shibuya, A. High Accuracy Geochemical Map Generation Method by a Spatial Autocorrelation-Based Mixture Interpolation Using Remote Sensing Data. Remote Sens. 2020, 12, 1991. https://doi.org/10.3390/rs12121991
Huang C, Shibuya A. High Accuracy Geochemical Map Generation Method by a Spatial Autocorrelation-Based Mixture Interpolation Using Remote Sensing Data. Remote Sensing. 2020; 12(12):1991. https://doi.org/10.3390/rs12121991
Chicago/Turabian StyleHuang, Chenhui, and Akinobu Shibuya. 2020. "High Accuracy Geochemical Map Generation Method by a Spatial Autocorrelation-Based Mixture Interpolation Using Remote Sensing Data" Remote Sensing 12, no. 12: 1991. https://doi.org/10.3390/rs12121991