Estimate of carbon stock in the soil via diffuse reflectance spectroscopy (vis/nir) air and orbital remote sensing

Faria, Ohana C. O.; Torres, Gilmar N.; Raimo, Luis A. D. L. Di; Couto, Eduardo G.

doi:10.1590/1983-21252023v36n320rc

Acessibilidade / Reportar erro

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Agricultural Engineering • Rev. Caatinga 36 (3) • Jul-Sep 2023 • https://doi.org/10.1590/1983-21252023v36n320rc copy

Estimate of carbon stock in the soil via diffuse reflectance spectroscopy (vis/nir) air and orbital remote sensing

Estimativa do estoque de carbono no solo via espectroscopia de reflectância difusa (vis/nir) sensoriamento remoto aéreo e orbital

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

Current procedures for determining soil organic carbon (SOC) content are costly, time-consuming, and generate polluting chemical waste. Therefore, developing new protocols using aerial and orbital remote sensing and diffuse reflectance spectroscopy (DRS) for digitally mapping the stock of soil organic carbon (CS) is essential for promoting actions of research and monitoring SOC in Brazilian soils. Given this, three areas of commercial plots in the region of the Middle North of Mato Grosso were studied, where sampling was carried out for the determination of SOC in the layer from 0 to 30 cm, evaluated by the dry combustion method and estimated through DRS in the visible to near

-infrared region - Vis-NIR-SWIR/350-2500 nm). To obtain the images by aerial remote sensing, the Carcará II^® Unmanned Aerial Vehicle was used, with a MicaSense^® multispectral camera (RGB + NIR + RedEdge) attached. The orbital sensors used were the Sentinel 2® and Planet® satellites. This study showed that soil carbon stock values could be predicted using different modeling approaches based on field and laboratory spectral measurements. Predictive models to estimate SOC can be established using remote and near sensing, thus allowing a better understanding of spatial patterns of SOC in crop fields.

Keywords
Precision agriculture; Carbon stock monitoring; Remote sensors

Farm	Clay	Sand	Silt	Organic Matter	BS	CEC	Ca	Mg	K	P	S
			-(%)---				---(cmol_c.dm^-3)--			--(mg.dm^-3)--
Capuaba	62	33	5	3.3	54	8.8	3.3	1.2	96	32	16
Palminha	53	38	9	3.0	57	7.1	2.8	0.6	70	27	15
Aterrado	21	72	7	1.9	69	5.6	2.8	0.7	41	28	7

Variable		Aterrado farm		Capuaba farm		Palminha farm		Regional
Variable		Crop	Reference	Crop	Reference	Crop	Reference	Regional
	Average	8.4	14.0	16.7	19.5	15.6	22.8	16.2
SOC	Minimum	5.7	9.2	9.9	12.3	8.1	16.5	10.3
(g kg^-1)	Maximum	12.8	21.1	21.9	43.6	19.3	26.9	24.3
	SD	1.6	6.3	2.4	12.0	2.5	4.8	4.9
	Average	1.4	1.2	1.1	1.2	1.1	1.0	1.2
Db (Mg m^-3)	Minimum	1.2	1.1	1.0	0.8	1.0	0.9	1.0
Db (Mg m^-3)	Maximum	1.5	1.3	1.2	1.4	1.3	1.1	1.3
	SD	0.1	0.1	0.1	0.2	0.1	0.1	0.1
	Average	29.7	50.8	58.0	61.5	47.9	69.7	52.9
CS	Minimum	20.0	29.3	34.5	46.3	24.9	49.2	34.0
(Mg ha^-1)	Maximum	45.0	82.4	75.9	98.1	59.4	90.1	75.1
	SD	5.7	27.9	8.4	18.9	7.8	15.1	14.0

			GPR model					SVM model
	Dataset	Statistical parameters	Cap	Pal	At	Reg	Ge	Cap	Pal	At	Reg	Ge
		R²	0.74	0.86	0.91	0.75	0.76	0.74	0.85	0.93	0.7	0.77
	Cal	RMSE	0.14	0.15	0.09	0.25	0.33	0.14	0.15	0.08	0.3	0.33
SOC (%)		RPIQ	1.39	2.74	2.13	2.44	2.43	1.38	2.71	2.48	2.4	2.43
SOC (%)	Val	_R2	0.84	0.49	0.67	0.61	0.66	0.87	0.22	0.31	0.6	0.67
	Val	RMSE	0.63	0.25	0.1	0.36	0.37	0.62	0.3	0.14	0.4	0.36
		RPIQ	0.26	0.55	2.21	1.78	2.25	0.26	0.46	1.61	1.8	2.27
		R²	0.92	0.82	0.94	0.73	0.76	0.94	0.8	0.99	0.7	0.76
	Cal	RMSE	1.75	1.56	0.78	2.53	3.19	1.52	1.71	0.3	2.6	3.23
SOC (g kg-¹)		RPIQ	1.01	1.96	3	2.49	2.51	1.16	1.79	7.72	2.4	2.47
SOC (g kg-¹)		_R2	0.15	0.07	0.34	0.63	0.69	0.17	0.15	0.29	0.6	0.68
	Val	RMSE	3.52	4.37	1.59	3.84	3.94	3.38	4.16	1.76	3.5	4.04
		RPIQ	1.18	0.53	1.14	1.62	2.22	1.23	0.56	1.03	1.8	2.16
		_R2	0.7	0.72	0.76	0.87	0.79	0.67	0.73	0.99	0.9	0.8
	Cal	RMSE	0.04	0.05	0.05	0.07	0.09	4.97	3.82	2.26	7.7	12.8
Db		RPIQ	3.05	2.14	2.92	4.73	3.57	1.85	2.04	2.36	2.2	1.74
Db		R²	0	0.07	0.26	0.78	0.71	0.04	0.05	0.8	0.4	0.15
	Val	RMSE	0.19	7.52	0.22	0.09	0.11	6.64	8.89	9.48	11	17.9
		RPIQ	0.57	1.15	0.42	3.81	2.85	0.88	0.97	0.91	1.2	1.16
		_R2	0.86	0.85	0.82	0.73	0.51	0.85	0.92	0.85	0.7	0.45
	Cal	RMSE	4.81	5.07	2.41	7.66	12.2	4.97	3.82	2.26	7.7	12.8
CS		RPIQ	1.91	1.53	2.22	2.24	1.82	1.85	2.04	2.36	2.2	1.74
CS		_R2	0.27	0.22	0.79	0.44	0.18	0.04	0.05	0.8	0.4	0.15
	Val	RMSE	4.36	7.52	9.34	10.8	17	6.64	8.89	9.48	11	17.9
		RPIQ	1.34	1.15	0.9	1.27	1.23	0.88	0.97	0.91	1.2	1.16
		_R2	0.93	0.9	0.72	0.5	0.49	0.97	0.92	0.49	0.6	0.67
	Cal	RMSE	0.07	0.12	0.09	0.35	0.48	0.04	0.11	0.13	0.31	0.38
SOC (%)		RPIQ	2.74	3.44	2.36	1.76	1.69	4.7	3.78	1.73	1.99	2.09
SOC (%)		R²	0.54	0.12	0.81	0.44	0.47	0.69	0.06	0.05	0.57	0.64
	Val	RMSE	0.74	0.44	0.19	0.42	0.46	0.66	0.37	0.41	0.37	0.38
		RPIQ	0.22	0.31	0.93	1.51	1.78	0.25	0.38	0.44	1.73	2.17
		MLR model							PLSR model
	Dataset	Statistical parameters	Cap	Pal	At	Reg	Ge	Cap	Pal	At	Reg	Ge
		R²	0.99	0.86	0.91	0.55	0.53	0.99	0.85	0.94	0.64	0.6
	Cal	RMSE	0.58	1.36	0.88	3.62	4.41	0.48	1.42	0.74	2.94	4.03
SOC (g kg-¹)		RPIQ	3.05	2.24	2.64	1.65	1.81	3.7	2.15	3.14	2.09	1.98
SOC (g kg-¹)		_R2	0.43	0.5	0.41	0.52	0.6	0.31	0.09	0.35	0.67	0.67
	Val	RMSE	5.28	3.81	1.72	3.19	4.46	4.02	4.81	1.87	3.3	4.04
		RPIQ	0.79	0.61	1.05	2.12	1.96	1.03	0.48	0.97	1.93	2.17
		_R2	0.72	0.59	0.31	0.76	0.59	0.55	0.45	0.46	0.82	0.68
	Cal	RMSE	2.25	3.64	2.65	9.91	14.73	6.42	6.68	2.32	9.67	14.67
Db		RPIQ	3.66	2.14	3.42	1.73	1.51	1.43	1.16	2.3	1.78	1.52
Db		R²	0.24	0.24	0.84	0.45	0.19	0.44	0.13	0.78	0.46	0.17
	Val	RMSE	11.58	8.9	7.43	10.62	16.6	4.1	7.93	8.4	10.6	17.02
		RPIQ	0.51	0.97	1.12	1.29	1.24	1.42	1.09	1.03	1.29	1.22
		_R2	0.96	0.92	0.81	0.53	0.25	0.73	0.73	0.82	0.55	0.25
	Cal	RMSE	2.25	3.64	2.65	9.91	14.73	6.42	6.68	2.32	9.67	14.67
CS		RPIQ	3.66	2.14	3.42	1.73	1.51	1.43	1.16	2.3	1.78	1.52
CS		R²	0.24	0.24	0.84	0.45	0.19	0.44	0.13	0.78	0.46	0.17
	Val	RMSE	11.58	8.9	7.43	10.62	16.6	4.1	7.93	8.4	10.6	17.02
		RPIQ	0.51	0.97	1.12	1.29	1.24	1.42	1.09	1.03	1.29	1.22

			GPR model				SVM model
	Data set	Statistical parameters	Cap	Pal	At	Cap	Pal	At
		2 R	0.16	0.25	0.81	0.79	0.23	1
	Cal	RMSE	0.23	0.25	0.11	0.11	0.24	0.02
		RPIQ	0.7	1.2	2.26	1.43	1.26	14.48
SOC (%)		2 R	0.08	0.17	0.45	0.28	0.11	0.48
	Val	RMSE	0.29	0.24	0.14	0.26	0.23	0.13
		RPIQ	0.41	1.09	0.84	0.46	1.14	0.92
		2 R	0.15	0.31	0.85	0.49	0.42	0.87
	Cal	RMSE	2.71	2.47	1.03	1.85	2.26	0.8
SOC (g kg^-1)		RPIQ	0.45	0.85	2.5	0.66	0.92	3.52
		2 R	0.55	0.04	0.32	0.03	0.38	0.35
	Val	RMSE	1.88	1.97	1.35	2.93	2.64	0.94
		RPIQ	1.05	2	0.68	0.68	1.5	0.69
		2 R	0.32	0.25	0.81	0.59	0.25	0.86
	Cal	RMSE	9.33	7.71	3.77	8.57	8.07	3.03
		RPIQ	0.39	1.21	2.3	0.42	1.16	2.86
CS		2 R	0.55	0.11	0.35	0.49	0.2	0.45
	Val	RMSE	7.58	7.84	4.49	5.42	5.73	4.35
		RPIQ	0.84	0.68	0.94	1.18	1.17	0.97
	MLR model					PLSR model
	Data set	Statistical parameters	Cap	Pal	At	Cap	Pal	At
		2 R	0.78	0.57	0.77	0.18	0.32	0.43
	Cal	RMSE	0.1	0.17	0.08	0.18	0.22	0.13
		RPIQ	1.55	1.8	2.97	0.89	1.33	1.89
SOC (%)		2 R	0.55	0.38	0.26	0.09	0.22	0.45
	Val	RMSE	0.23	0.24	0.31	0.29	0.21	0.14
		RPIQ	0.51	1.09	0.38	0.41	1.22	0.85
		2 R	0.92	0.81	0.67	0.11	0.26	0.87
	Cal	RMSE	0.68	1.14	1	2.11	2.22	0.8
SOC (g kg^-1)		RPIQ	2.67	1.84	2.79	0.94	0.94	3.52
		2 R	0.01	0.28	0.04	0.31	0.18	0.35
	Val	RMSE	3.32	3.64	2.7	2.91	2.18	0.94
		RPIQ	0.37	1.09	0.24	1.2	1.82	0.69
		2 R	0.89	0.69	0.34	0.62	0.24	0.36
	Cal	RMSE	3.07	4.47	4.71	5.84	6.97	4.64
		RPIQ	1.18	2.09	1.84	0.62	1.34	1.87
CS		2 R	0.21	0.05	0.37	0.61	0.19	0.82
	Val	RMSE	10.38	7.92	4.81	9.07	6.27	2.71
		RPIQ	0.61	0.85	0.87	0.7	1.07	1.55

		GPR model				SVM model
	Data set	Statistical parameters	Cap	Pal	At	Cap	Pal	At
		_R2	0.97	0.19	0.76	0.97	0.22	0.8
	Cal	RMSE	0.03	0.26	0.08	0.03	0.27	0.07
SOC (%)		RPIQ	5.78	1.13	2.01	4.35	1.09	2.33
SOC (%)		R²	0.44	0.08	0.15	0.15	0.09	0.12
	Val	RMSE	0.7	0.21	0.19	0.39	0.22	0.22
		RPIQ	0.63	1.36	0.93	1.13	1.26	0.81
		_R2	0.64	0.98	0.77	0.81	0.3	0.75
	Cal	RMSE	1.86	1.11	1.12	1.36	2.19	1.06
SOC (g kg^-1)		RPIQ	1.29	2.65	2.2	1.76	1.34	2.32
SOC (g kg^-1)		R²	0.01	0.11	0.66	0.03	0.08	0.5
	Val	RMSE	1.77	3.33	0.94	1.25	3.34	0.93
		RPIQ	0.39	1.2	0.99	0.55	1.19	0.99
		_R2	0.97	0.15	0.75	0.15	0.17	0.9
	Cal	RMSE	3.6	8.7	3.23	6.8	8.94	2.29
		RPIQ	1.57	1.59	1.85	0.83	1.54	2.62
		R²	0.28	0.15	0.05	0.62	0.24	0.12
	Val	RMSE	10.68	4.05	6.25	11.12	3.66	6.5
		RPIQ	0.42	0.56	0.7	0.4	0.62	0.67
		MLR model				PLSR model
	Data set	Statistical parameters	Cap	Pal	At	Cap	Pal	At
		_R2	0.51	0.18	0.61	0.69	0.16	0.67
	Cal	RMSE	0.1	0.24	0.08	0.08	0.24	0.08
SOC (%)		RPIQ	1.49	1.24	2.07	1.87	1.23	2.25
SOC (%)		R²	0.4	0.07	0.23	0.5	0.02	0.17
	Val	RMSE	1.29	0.21	0.2	0.51	0.21	0.18
		RPIQ	0.34	1.35	0.89	0.85	1.35	1.01
		_R2	0.58	0.43	0.51	0.38	0.13	0.39
	Cal	RMSE	1.76	1.59	1.21	2.13	1.96	1.34
SOC (g kg^-1)		RPIQ	1.36	1.84	2.05	1.13	1.5	1.84
SOC (g kg^-1)		R²	0	0.01	0.39	0.03	0.15	0.52
	Val	RMSE	2.34	3.85	1.11	2.51	3.4	0.89
		RPIQ	0.29	1.04	0.84	0.27	1.17	1.03
		R²	0.39	0.11	0.61	0.49	0.13	0.67
	Cal	RMSE	5.33	8.23	3.17	4.85	8.12	2.92
CS		RPIQ	1.06	1.68	1.89	1.16	1.7	2.05
CS		_R2	0.62	0.11	0.2	0.28	0.01	0.1
	Val	RMSE	9.11	4.41	6.05	9.47	5.05	5.76
		RPIQ	0.49	0.52	0.72	0.47	0.45	0.76

				GPR model		SVM model				MLR model
	Data set	SP	Cap	Pal	At	Reg	Cap	Pal	At	Reg	Cap	Pal	At	Reg
		R²	0.23	0.54	0.41	0.66	0.13	0.56	0.41	0.71	0	0.0	0.0	0.0
	Cal	RMSE	0.19	14	1.54	0.24	0.2	0.14	0.11	0.23	0.22	2	0.14	0.43
SOC (%)		RPIQ	0.53	1.4	1.12	3.32	0.5	1.49	1.84	3.4	0.47	1.01	1.43	1.85
SOC (%)		_R2	0.0	0.05	0.05	0.23	0.01	0.04	0.02	0.23	0.0	0.0	0.0	0.0
	Val	RMSE	0.42	0.34	3.6	0.39	0.39	0.36	0.27	0.42	0.3	0.35	0.23	0.47
		RPIQ	0.91	0.98	0.48	0.7	0.97	0.92	0.84	0.65	1.26	0.94	0.97	0.58
		R²	0.41	0.45	0.51	0.59	0.41	0.34	0.57	0.58	0.0	0.0	0.12	0.04
	Cal	RMSE	1.54	1.94	1.27	2.88	1.64	2.12	1.19	2.94	1.98	2.59	1.66	4.42
SOC (g.kg-¹)		RPIQ	1.12	1.85	1.43	2.87	1.05	1.69	1.53	2.81	0.86	1.39	1.09	1.87
SOC (g.kg-¹)		_R2	0.05	0.03	0.02	0.75	0.01	0.34	0.05	0.78	0	0	0.1	0.03
	Val	RMSE	3.6	2.53	2.83	2.37	3.53	1.78	3.43	2.06	3.28	2.23	1.82	3.92
		RPIQ	0.48	0.7	0.57	2.9	0.49	1	0.47	3.34	0.53	0.79	0.9	1.75
		R²	0.25	0.45	0.48	0.47	0.2	0.43	0.45	0.42	0	0.12	0.11	0.05
	Cal	RMSE	8.45	5.82	5.67	8.73	8.87	5.9	5.8	9.15	9.69	7.31	7.35	11.69
CS		RPIQ	0.98	1.79	0.93	2.19	0.93	1.77	0.91	2.09	0.85	1.43	0.72	1.63
CS		_R2	0.16	0.02	0.02	0.49	0.78	0.2	0.2	0.45	0	0.21	0.03	0.27
	Val	RMSE	51.4	8.78	8.37	8.09	25.57	6.85	11.7	8.44	2.56	8.65	5.52	10.42
		RPIQ	0.06	0.8	0.95	1.79	25.57	1.03	0.68	1.72	1.22	0.81	1.44	1.39

Universidade Federal Rural do Semi-Árido Avenida Francisco Mota, número 572, Bairro Presidente Costa e Silva, Cep: 5962-5900, Telefone: 55 (84) 3317-8297 - Mossoró - RN - Brazil
E-mail: caatinga@ufersa.edu.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Conflict of interest: The authors declare no conflict of interest related to the publication of this manuscript.