Machine learning methods for estimation the indicators of phosphogypsum influence in soil

Purpose

The full understanding of the effect of mineral waste-based fertilizer in soil is still unrelieved, because of the extreme complex chemical composition and plethora of their action pathways. The purposes of this paper is to quantify the input of PG into the soil ecosystem process, considering the direct effects of PG as a whole on soil environment using of a plethora of chemical, toxicological, and biological tests.

Materials and methods

Greenhouse experiment includes different PG doses (0, 1%, 3%, 7.5%, 15%, 25%, and 40%) and two-time collection points after treatments—7 and 28 days. For each treatment and each time collection point, we measure (i) soil pH, bioavailable (H₂0 and NH₄COOH-extractable) element content (S, P, K, Na, Mg, Ca, Fe, Zn, Sr, Ba, F); (ii) soil enzyme activities—dehydrogenase, urease, acid phosphatase, FDA; (iii) soil CO₂ respiration activity with and without glucose addition; (iv) Eisenia fetida, Sinapis alba, and Avena sativa responses. Finally, we combine the ordinary chemical, toxicology, and biological measuring of soil properties with state-of-the-art mathematical analysis, namely (i) support vector machines (used for prediction), (ii) mutual information test (variable importance tasks), (iii) t-SNE and LLE algorithms (used for unsupervised classification).

Results and discussion

The results show similarity between the 0%, 1%, and 3% PG treatments in all collection times based on the toxicological and biological properties. Beyond 7.5% PG, some biological test was significantly inhibited in response to trace element stress. Among all tested parameters, soil urease activities, soil respiration activities after glucose addition, S. alba root lengths, and E. fetida survival rates show sensitivity to PG addition. Furthermore, the machine learning algorithms revealed that only several elements (mobile and water-soluble forms of Ca, Ba, Sr, S, and Na; water-soluble F) could be responsible to elevated soil toxicity for those indicators. SVR models were able to predict soil biological and ecotoxicity properties, and increasing numbers of randomly selected training examples from 50 to 90% of initial experimental data significantly improved model performance.

Conclusions

At this study, we demonstrate benefits of unsupervised machine learning methods for investigating toxicity of man-made substances in soil that can be further applied to risk assessments of various toxins, which are of significant interest to environmental protection.

     

Impact of fertilization on cover crops and microbial community on a bauxite‐mined soil undergoing reclamation

The reclamation of bauxite‐mined areas can be favored by the application of organic and/or chemical fertilization to restore the vegetation. Otherwise, the impact of fertilizations on soil microbiota or plant–microbe interactions as land reclamation progresses is less understood. To address this issue, we evaluated the impact of organic and chemical fertilization on plants and soil microbial community within the first 36 months of land reclamation in a bauxite‐mined site. The experiment was set up according to a split‐plot design in which the main plots received fertilizations [non‐fertilized control (NF), chemical fertilization (CF; NPK and rock phosphate), organic fertilization (OF; poultry litter), and CF+OF combined], and the subplots received cover crops [no cover crops (NC), grass (B; Brachiaria), legume (S, Stylosanthes), and B+S combined]. Cover crops biomass yield was assessed annually with five field campaigns per year. We used phospholipid fatty acids (PLFAs) to infer the impacts of mining and restoration practices on actinobacteria, Gram‐negative and Gram‐positive bacteria, arbuscular mycorrhizal fungi, and fungi. Accordingly, PLFAs were determined before bauxite mining (pre‐mining), six months after topsoil reconfiguration (post‐mining), and after 14 and 36 months following the application of the fertilizations and cover crops. PLFAs results indicated that in post‐mining, the living microbiota was significantly lower than in pre‐mining. Cover crops biomass yield was highest for B and B+S fertilized with CF+OF at 14 and 36 months. Both parametric and non‐parametric statistics showed a temporal variation in the response of living microbes to the treatments applied. After 14 months, living microbes showed greatest response to OF, while at 36 months their response was strongest in the treatments with highest plant biomass production (B and B+S). These results suggest that in the early stages of land reclamation, living microbial biomass benefit the most from organic fertilizers. As this initial boost decline, living microbes are more likely to thrive in areas undergoing reclamation where they can develop synergistic interactions with plants.     

Contamination history of suspended river sediments accumulated in oxbow lakes over the last 25 years

Background, aims, and scope  Embankment of meandering river systems in many industrial areas results in the formation of artificial oxbow lakes that may act as perennial or intermittent traps for river sediments. Their deposits can be dated using a combination of historical and stratigraphic data, providing a good means to study historical records of contamination transported by rivers. Contamination history over the last few decades is of special significance for Central and Eastern Europe as it can reflect high pollutant levels in the second half of the twentieth century and the subsequent improvement after the fall of the Iron Curtain. The purpose of this study was to investigate recent sediments of an oxbow lake of the Morava River, Czech Republic, their stratigraphic records, sediment architecture, and history of contamination. Materials and methods  Seven ground-penetrating radar (GPR) profiles and three sediment cores up to 4 m deep were studied. The stratigraphy of the cores was inferred from visible-light spectrophotometry, X-ray radiography, grain size analysis, and semiquantitative modal analysis of sandy fractions. The sediments were dated using the ¹³⁷Cs mass activity and combinations of stratigraphic and historical data. The cores were sampled for concentrations of heavy metals and persistent organic pollutants. Wet sampled, lyophilized, and sieved sediments were extracted and analyzed for heavy metals by inductively coupled plasma mass spectrometry (ICP-MS) of aqua regia leachate and for persistent organic pollutants by gas chromatography (GC-ECD and GC-MS). Results  Three distinct sedimentary sequences (S1, S2, and S3) were identified. The basal sequence S1 represents river channel sediments deposited before the formation of the oxbow lake, most likely before the 1930s. The boundary between the S1 and S2 sequence correlates with the level of sediment dredging from 1981 evidenced from historical data. The overlying sequences S2 and S3 represent a postdredging sediment wedge, which progrades into the lake. ¹³⁷Cs dating revealed a distinct Chernobyl 1986 peak at ∼150 cm depth inferring sedimentation rates up to 7.7 cm/year. Sediment contamination abruptly increased from the pre-1930s deposits to the post-1981 deposits. The concentration levels increased two to five times for Pb, As, Zn, and Cu, about 10 to 15 times for Cr, Sb, and Hg, up to 34 times for Cd, and 25 to 67 times for DDTs, PCBs, and PAHs. The concentrations of most contaminants remained approximately constant until the late 1980s when they started to decrease slowly. The decreasing trends were most prominent for heavy metals and anthracene, less prominent for DDTs, and almost absent for PCBs and most PAHs. Discussion  Different temporal and spatial patterns for various contaminants make it possible to distinguish between anthropogenic point sources from local industry (anthracene, Cr, Cd), possible diffuse sources (most PAHs), and geological background (V, Co, Ni, and Mo). The observed recent trends in heavy metal and persistent organic pollutant contamination are generally consistent with data from other Central European rivers. The roughly balanced contamination levels in sediments from the lake and the adjacent river channel suggest that the oxbow lake deposits reflect immediate levels of the contamination bound to suspended particulate matter passing through the river. Conclusions  The investigated oxbow lake accumulated suspended sediment from Morava River, developing a thick sedimentary body. The sediments offer a good time framework to study historical contamination of the river on a decade time scale. Continuous contamination trends can be traced back to the early 1980s. The results show that stratigraphic analysis of cores has a good potential for identification of uninterrupted historical trends and unconformities, e.g., due to dredging. Recommendations and perspectives  Oxbow lakes may provide an alternative to floodplains and reservoir deposits when studying river contamination history. High sedimentation rates of up to 7.7 cm/year offer a very good stratigraphic resolution, making it possible to study contamination patterns on annual or even seasonal time scales. On the other hand, oxbow lakes may represent long-term contamination stores, which are unlikely to be redistributed by river erosion and, hence, may possess significant environmental risks for the farther future.     

Nitrogen mineralization in marsh meadows in relation to soil organic matter content and watertable level

The objective of the present study was to asses the effect of watertable level on N mineralization in a Histosol and a Humic Gleysol profile under natural meadows in Ljubljana marsh, Slovenia. The two soils differ significantly in organic matter content (27—40 % in Histosol and 14—20 % in Humic Gleysol) but not in C : N ratio (13—20) and pH (6.5—7.0). For each soil, the watertable was maintained at two levels (above or below 50 cm from the soil surface) for approximately one year. The four main plots, according to soil carbon content and watertable level were divided into 4 subplots, according to 4 fertilization treatments (unfertilized control, PK, PK + 50 kg N ha^—1, PK + 3 × 50 kg N ha^—1). Net N mineralization in unfertilized subplots was estimated from indices of N mineralization obtained by incubation of soil samples in the laboratory and by seasonal dynamics of mineral N content in the field. Annual uptake of N in herbage under the 4 fertilization treatments was also measured. Total mineral N content in topsoil was 20—80 % higher in Histosol than in Humic Gleysol. Similarly, aerobic N mineralization potentials along the entire soil profile (0—90 cm) were 20—130 % higher in Histosol than in Humic Gleysol. By contrast, anaerobic N mineralization potentials in subsoil were 10—60 % lower in Histosol than in Humic Gleysol. Both, aerobic and anaerobic N mineralization potentials strongly depended on watertable levels at sampling time. Seasonal dynamics of soil mineral N content as well as N mineralization potentials indicated that the N mineralization in the Histosol could be 10—40 % higher at low than at high watertable level. In the Humic Gleysol the N mineralization could be 10—100 % higher at high watertable level. Higher N availability in Histosol at low watertable and in Humic Gleysol at high watertable was also reflected in higher N uptake in herbage. These results indicate that N mineralization in Histosol and Humic Gleysol, was proportional to soil organic matter content, whereas in both soils, higher N mineralization rates can be expected at watertable levels between 40 and 60 cm below the soil surface, than at higher/lower watertable levels.     

Boom‐bust like dynamics of invasive black bullhead (Ameiurus melas) in Lake Sava (Serbia)

The life‐history traits and population dynamics of the black bullhead Ameiurus melas (Rafinesque) were studied monthly from August 2009 to December 2012 in Lake Sava in Belgrade (Serbia). This period included a mass mortality event in May 2011 and subsequent population recovery during 2012. Available evidence suggests the mass mortality was caused by European catfish virus that only affected the black bullhead. Changes in key life‐history traits and the recovery potential (increase in catch‐per‐unit‐effort, doubling of young‐of‐the‐year to adult ratio, earlier maturation and increased fecundity) following the mass mortality event suggest population resilience typical of a recurring boom‐bust pattern and demonstrate the species’ capacity to recover rapidly and re‐establish following disturbance. Repeated systematic mass removals of both juveniles and adults (using fyke nets) will be required if future fishery management measures aimed at limiting the population size are to be effective.     

An alarming tendency towards freshwater mussel misidentification in scientific works may bias endangered and invasive species management

相似文献   

Pathogenicity of fungi associated with ash dieback symptoms of one‐year‐old Fraxinus excelsior in Montenegro

In addition to Hymenoscyphus fraxineus, two fungi identified as Diaporthe eres aff. and Fusarium sambucinum aff. were also isolated from necrotic bark lesions on declining one‐year‐old Fraxinus excelsior in a forest stand in Montenegro. To examine their involvement in ash decline, a pathogenicity test was performed using under bark inoculations on one‐year‐old Fraxinus excelsior. Hymenoscyphus fraxineus was included as comparison. All three fungal species proved highly pathogenic towards one‐year‐old seedlings although lesion sizes differed significantly between the different species. Hymenoscyphus fraxineus was most aggressive, followed by F. sambucinum aff., while D. eres aff. caused the smallest lesions. This study demonstrates for the first time the ability of isolates in the D. eres and F. sambucinum species complexes to cause decline on one‐year‐old common ash seedlings.     

Effect of faunation on ruminal solubility and liver content of copper in sheep fed low or high copper diets

Naturally fauna-free crossbred rams approximately 5 mo old, 49.3 +/- 1.88 kg live weight, were allotted to four groups. Two groups remained fauna-free, whereas the other two groups were faunated. A diet based on corn silage-soybean meal containing either a low or a high Cu concentration (7 or 15 micrograms/g dry matter) was fed ad libitum to the two groups for 197 d. The rams were killed at the end of the experiment and livers and ruminal contents were analyzed for Cu. Faunated sheep grew faster (P less than .01) and as a group consumed 1.53 kg diet dry matter daily compared with 1.34 for the fauna-free sheep. Faunated sheep had lower final liver Cu concentrations (P less than .01), total liver Cu per unit of Cu intake and ruminal Cu solubilities (P less than .01) when fed both low and high Cu diets. The relative hepatic Cu uptake was higher by rams fed the low Cu than by those fed the high Cu diet. The results are interpreted to show that faunation decreased the availability of Cu in sheep fed a corn silage-soybean meal diet. The magnitude of the effect was independent of the Cu concentration in the diet.