Reflectance spectroscopy in the prediction of soil organic carbon associated with humic substances

Ribeiro, Sharon Gomes; Oliveira, Marcio Regys Rabelo de; Lopes, Letícia Machado; Costa, Mirian Cristina Gomes; Toma, Raul Shiso; Araújo, Isabel Cristina da Silva; Moreira, Luis Clenio Jario; Teixeira, Adunias dos Santos

doi:10.36783/18069657rbcs20220143

Acessibilidade / Reportar erro

Brasil

Revista Brasileira de Ciência do Solo

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Division – Soil in Space and Time, Commission – Pedometrics • Rev. Bras. Ciênc. Solo 47 • 2023 • https://doi.org/10.36783/18069657rbcs20220143 copy

Reflectance spectroscopy in the prediction of soil organic carbon associated with humic substances

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

Understanding organic carbon and predominant humic fractions in the soil allows contributes to soil quality management. Conventional fractionation techniques require time, excessive sampling, and high maintenance costs. In this study, predictive models for organic carbon in humic substances (HS) were evaluated using hyperspectral data as an alternative to chemical fractionation and quantification by wet digestion. Twenty-nine samples of Neossolos Flúvicos (Fluvents) - A1, and 36 samples of Cambissolos (Inceptisols) - A2 were used. The samples were also analyzed jointly, creating a third sample group - A1&A2. Untransformed spectral reflectance factors were obtained using the FieldSpec Pro FR 3 hyperspectral sensor (350–2500 nm). Pre-processing techniques were employed, including Savitzky–Golay smoothing and first- and second-order derivative analysis. After selecting variables using the Backward method, which removes spectral variables that are not statistically significant for the regression. Estimation models were built by Principal Components Regression (PCR) and Partial Least Squares Regression (PLSR). The spectral data were evaluated individually for soil classes A1 and A2, and jointly for A1&A2. The PLSR was more efficient than PCR, especially for the estimation models that used the first derivative of reflectance employing the three sample groups. For samples of A1, the best estimate was seen for humic acid (RPD = 6.09) and humin (RPD = 2.38); for A2, the best models estimated the OC in fulvic acid (RPD = 2.35) and humin (RPD = 2.51); and for the joint spectral data (A1&A2), the prediction was robust for humin only (RPD = 2.01). The most representative wavelengths were observed using the first derivative with PLSR and PCR, centred on the region between 1600 and 1800 nm. The first-derivative of reflectance calculated more-robust predictive models using PLSR than PCR. The best predictions occurred for organic carbon associated with humic acid in Neossolos Flúvicos, with fulvic acid in Cambissolos, and with humin in both soil classes.

Keywords
spectroradiometry; pedometrics; organic matter

Sample	Humic substances	Spectral data	Best PCR prediction models for C-HS (normalised)	R²_adj.
A1	HA	First Derivative (ρ’)	0.475 - 2106.56 (ρ’_832nm) + 92.94 (ρ’_2317nm) – 230.83 (ρ’_371nm) + 2243.5 (ρ’_912nm) – 1789.59 (ρ’_901nm) + 2900.01 (ρ’_843nm) – 1461 (ρ’ _1223nm) + 323.5 (ρ’_1606nm) – 987.23 (ρ’_1661nm) – 197.34 (ρ’_2331nm) + 360.52 (ρ’_1821nm)	0.95
A1	UM	First Derivative (ρ’)	-0.786 – 2493.96 (ρ’_986nm) – 1676.08 (ρ’_1669nm) – 996.82 (ρ’_1851nm) + 3227.71 (ρ’_1229nm) + 1501.33 (ρ’_842nm) + 455.81 (ρ’_1610nm) + 73.62 (ρ’_1665nm) + 1371.01 (ρ’_1132nm) + 96.89 (ρ’_1231nm) - 340.79 (ρ’_1491nm)	0.80
A2	FA	First Derivative (ρ’)	0.430 - 3950.04 (ρ’_995nm) – 1463.14 (ρ’_1721nm) – 31.01 (ρ’_368nm) + 2226.65 (ρ’_987nm) + 118.42 (ρ’_357nm) – 647.41 (ρ’_1700nm) - 196.15 (ρ’_382nm) + 1136.86 (ρ’_1017nm) – 646.25 (ρ’_1726nm) – 486.81 (ρ’_1606nm) + 59.99 (ρ’_1737nm) – 311.95 (ρ’_431nm)	0.79
A2	HUM	First Derivative (ρ’)	0.329 – 1384.06 (ρ’_1719nm) + 1384.47 (ρ’_1023nm) + 1050.97 (ρ’_1813nm) + 1311.44 (ρ’_1263nm) + 162.64 (ρ’_1606nm) – 648.93 (ρ’_628nm) – 801.13 (ρ’_1044nm) + 834.88 (ρ’_2312nm) – 211.00 (ρ’_371nm) + 40.66 (ρ’_361nm) + 533.06 (ρ’_987nm)	0.84

Sample	Humic substances.	Spectral data	Best PLSR prediction models for C-HS (normalised)	R²_adj.
A1	HA	First Derivative (ρ’)	0.264 - 1553.99 (ρ’_832nm) + 285.72 (ρ’_2317nm) - 298.36 (ρ’_371nm) + 2906.19 (ρ’_912nm) - 1555.22 (ρ’_901nm) + 1897.48 (ρ’_843nm) - 1763.80 (ρ’_1223nm) + 864.44 (ρ’_1606nm) - 387.49 (ρ’_1661nm) - 252.65 (ρ’_2331nm) - 1.46 (ρ’_1821nm)	0.97
A1	HUM	First Derivative (ρ’)	- 0.789 - 2460.43 (ρ’_986nm) - 1678.28 (ρ’_1669nm) - 966.92 (ρ’_1851nm) + 2986.89 (ρ’_1229nm) + 1535.40 (ρ’_842nm) + 532.88 (ρ’_1610nm) + 60.59 (ρ’_1665nm) + 1383.32 (ρ’_1132nm) + 161.16 (ρ’_1231nm) - 281.35 (ρ’_1491nm)	0.80
A2	FA	First Derivative (ρ’)	0.413 - 3502.33 (ρ’_995nm) -1551.79 (ρ’_1721nm) - 570.99 (ρ’_368nm) + 1913.60 (ρ’_987nm) + 180.04 (ρ’_357nm) - 563.85 (ρ’_1700nm) -232.96 (ρ’_382nm) + 798.81 (ρ’_1017nm) - 852.76 (ρ’_1726nm) -111.49 (ρ’_1606nm) + 113.29 (ρ’_1737nm) - 198.84 (ρ’_431nm)	0.78
A2	HUM	First Derivative (ρ’)	0.369 - 1361.95 (ρ’_1719nm) + 801.05 (ρ’_1023nm) + 855.16 (ρ’_1813nm) + 1453.15 (ρ’_1263nm) + 389.22 (ρ’_1606nm) - 1002.90 (ρ’_628nm) - 453.46 (ρ’_1044nm) + 869.54 (ρ’_2312nm) -253.17 (ρ’_371nm) - 39.20 (ρ’_361nm) + 940.23 (ρ’_987nm)	0.82
A1&A2	HUM	First Derivative (ρ’)	0.562 + 35.91 (ρ’_1051nm) + 591.03 (ρ’_1608nm) - 282.31 (ρ’_981nm) - 393.34 (ρ’_966nm) + 665.06 (ρ’_1834nm) + 257.55 (ρ’_1828nm) + 54.19 (ρ’_1742nm) + 287.71 (ρ’_1853nm) - 486.48 (ρ’_1399nm) - 32.39 (ρ’_1462nm) + 90.44 (ρ’_388nm) - 296.36 (ρ’_1703nm) - 589.75 (ρ’_1976nm)	0.85

Sociedade Brasileira de Ciência do Solo Secretaria Executiva , Caixa Postal 231, 36570-000 Viçosa MG Brasil, Tel.: (55 31) 3899 2471 - Viçosa - MG - Brazil
E-mail: sbcs@ufv.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] *Corresponding author: E-mail: sharonribeiro@alu.ufc.br