Search

China is the largest coal consumer in the world. The massive exploitation and utilization of coal resources have resulted in serious problems of heavy metal pollution and environmental contamination, such as soil degradation, water pollution, crop damage, and even threatening human lives. Therefore, monitoring soil heavy metal pollution quickly and in real time is an urgent task at present. This research not only formulated a new preprocessing method enlightened by few-shot learning for soil hyperspectral data but also combined it with other soil-related auxiliary information to extract effective information from the soil hyperspectrum, at the end of which different regression methods were adopted to predict soil heavy metal contamination. This test used 168 actual soil samples from the Eastern Junggar coalfield in Xinjiang for verification. Since copper in the soil is a trace element and the corresponding spectral characteristics are affected by other impurities, improper use of hyperspectral preprocessing methods may introduce interference information or may delete useful information, which makes the model effect unsatisfied. To effectively address the above-mentioned problems, the preprocessing method of this experiment combined the second-order differential derivation, and the data enhancement (DA) method together with the addition of auxiliary information to allow more effective features to be entered into the model. Next, the attentive interpretable tabular learning (TabNet) model was improved in three different ways using the original TabNet model and three improved TabNet models to create regression models. One of the improved TabNet models had the best effect, with a list of the top 30 features according to the degree of importance. Meanwhile, the regression prediction of Cu content using four different convolutional neural networks (CNNs) revealed that the model with the residual block was the strongest and slightly outperformed the improved TabNet model, but lacked interpretation of the input data. Besides, this experiment also employed different preprocessing methods for regression prediction on various models and found that the traditional preprocessing methods performed best in traditional regression models [e.g., partial least square regression (PLSR)] and underperformed in deep learning models. The selected optimal model was compared with PLSR and CNN models. The results indicated that both the improved TabNet model and the improved CNN model had better performance using the new preprocessing approach proposed in this article, with improved TabNet yielding a coefficient of determination ( $\text{R}^{2}$ ), root-mean-square error (RMSE), and the ratio of performance to interquartile range (RPIQ) of 0.94, 1.341, and 4.474, respectively. The improved CNN model had a coefficient of determination of 0.942, an RMSE of 1.324, and an interquartile range of 4.531 in the test dataset.

[Resumo] A viticultura é unha actividade agrícola en expansión en moitas partes do mundo, xa que é un cultivo económicamente rentable. Nembargantes, o cultivo da vide xenera unha serie de problemas ambientáis, dos que algúns dos principais están relacionadas coa degradación de solos. Unha das causas máis importantes desta degradación é a acumulación de cobre no solo procedente da aplicación de funxicidas. Aínda que o Cu é pouco móbil en solos, pode mobilizarse baixo determinadas condicións. Neste estudo, analizouse o efecto da adición dunha emenda correctora do pH na mobilización de cobre nun solo de viñedo do Ribeiro no que o nivel deste metal excedeu os niveis naturais. Mostras deste solo empaquetáronse en columnas que foron lixiviadas en condicións de saturación con carga hidráulica constante. Os tratamentos considerados incluíron a aplicación de CaO en catro doses: 0 (control), 3.3 (recomendada para subir o pH do solo a 6.5), 14.3, e 46.6 g/kg de solo. A concentración de cobre, ademais da condutividade eléctrica, o pH e o volume dos lixiviados, mediuse a distintos tempos de saída. A condutividade eléctrica dos lixiviados aumentou ao aumentar a dose de cal, debido á liberación de electrólitos á solución do solo pola rápida disolución de cal. Así mesmo, a aplicación de cal aumentou o pH do solo de forma significativa, pero foi especialmente elevada con aplicacións de CaO por riba de 14g/kg, cando o pH sobrepasou valores de 12. Estes valores de pH poden resultar nun perxuicio para os organismos do solo e para a dispoñibilidade de nutrintes. O aumento do pH debeuse á rápida disociación do CaO que liberou gran cantidade de O-2 libres para formar OH-. A mobilidade do cobre aumentou lixeiramente cunha aplicación de CaO de 3.3g/kg, e de forma moito máis acusada con aplicacións por riba de 14g/kg. Este comportamento é similar ó do pH, o que suxire que están directamente relacionados. Á vista destes resultados, coa dose de cal recomendada para corrixir o pH deste solo a 6.5, non parece existir risco de mobilización de Cu. Nembargantes, con aplicacións superiores a esta, produciríase un aumento de pH moi acusado que conlevaría a mobilización de certas formas de Cu acumulado tras as aplicacións de fitosanitarios. Por isto, un forte encalado cunha dose de 14.3g/kg en condicións controladas de recollida de lixiviados pode ser considerado como un posible método para eliminar altas concentracións de este metal no solo. [Abstract]Viticulture is one of the agricultural activities in expansión in many parts of the World, since grapevines are a profitable crop. Nevertheless, vine cultivation results in a series of environmental problems, some of which are related to soil degradation. One of the most important causes of this degradation is Copper accumulation in the soil, as a side effect of the application of fungicides. Although Cu is poorly mobile in soils, it can be mobilized under certain conditions. In the present work, the effect of adding a pH amendment to the soil on Cu mobility was analysed in a vineyard soil from the Ribeiro region. This soil showed a Cu concentration that exceeded the natural values that could be expected in the soils under granitic parent materials. Soil samples were packed in columns and leached under saturated conditions with a constant hydraulic head. The treatments included CaO additions to the soil at four rates: 0 (control), 3.3 (lime required to increase the pH of this soil to 6.5), 14.3, and 46.6 g/kg soil. Copper concentration, in addition to electrical conductivity, pH and leachate volume, was measured at different times during each leaching run. Leachate electrical conductivity increased with increasing lime application rate, due to the release of electrolytes to the soil solution when lime is dissolved. Also, lime application significantly increased pH of the leachate, but the increase was especially acute with lime applications above 14g/kg, when pH reached values of 12. These pH values can be harmful for soil organisms and for nutrient availability. The increase of pH with the addition of lime was a result of the rapid disociation of CaO, which released a high amount of O-2 available to form OH-. Copper mobility increased slightly with a lime application rate of 3.3 g/kg, and much more acutely with application rates above 14 g/kg. Copper behaviour was similar to the one of pH, which suggests a direct relation between both. In light of these results, adding the lime required to rise the pH of this soil to 6.5 does not result in a risk of Cu mobilization. However, higher application rates can result in an acute pH rise and the consequent mobilization of some forms of Cu that accumulated in the soil after the application of fungicides. Thus, liming with an application rate of 14.3 g/kg under controlled conditions in which the leachate can be collected can be considered as a viable method to remove high concentrations of Cu from the soil. Traballo fin de grao (UDC.CIE). Bioloxía. Curso 2015/2016