Effects of cover crop growth and decomposition on the distribution of aggregate size fractions and soil microbial carbon dynamics

Although the effects of cover crops (CC) on various soil parameters have been fully investigated, less is known about the impacts at different stages in CC cultivation. The objective of this study was to quantify the influence of CC cultivation stages and residue placement on aggregates and microbial carbon (C_mic). Additionally, the influence of residue location and crop species on CO₂ emissions and leached mineralized nitrogen (N_min) during the plant degradation period was also investigated. Within an incubation experiment, four CC species were sown in soil columns, with additional columns being kept plant‐free. After plant growth, the columns were frozen (as occurs in winter under field conditions) and then incubated with the plant material either incorporated or surface‐applied. With CC, concentrations of large and medium macroaggregates were twice that of the fallow, confirming positive effects of root growth. Freezing led to a decrease in these aggregate size classes. In the subsequent incubation, the large macroaggregates decreased far more in the samples with CC than in the fallow, leading to similar aggregate size distributions. No difference in C_mic concentration was found among the CC cultivation stages. CO₂ emissions were roughly equivalent to the carbon amounts added as plant residues. Comparison of columns with incorporated or surface‐applied residues indicated no consistent pattern of aggregate distribution, CO₂ emission or C_mic and N_min concentrations. Our results suggest that positive effects of CC cultivation are only short term and that a large amount of organic material in the soil could have a greater influence than CC cultivation.     

Modelling of sorption experiments and seepage data of an Amazonian Ultisol subsoil under cropping fallow

The results of physico-chemical investigations of an Ultisol subsoil under a 2-year old fallow in eastern Amazonia are presented. Subsoil chemistry was studied using 4 different approaches: i) concentrations of H, Na, K, Ca, Mg, Mn, Al, and Fe in seepage water were measured under field conditions, ii) the equilibrium soil chemistry was studied in sequential batch experiments where the soil was treated with different solutions, iii) results of batch experiments were simulated with a chemical equilibrium model, and iv) the seepage data were calculated using selectivity coefficients obtained by modelling the batch experiments. The model included multiple cation exchange, precipitation/dissolution of Al(OH)₃ and inorganic complexation. Cation selectivity coefficients were pK_x/Ca^sel: X = Na: 0.3, K: 0.8, Mg: ?0.1, and Al: 0.4. The amount of cations sorbed ranged from ?0.2 to 2.0 (K), ?0.7 to 2.3 (Mg), ?1.6 to 1.8 (Ca), ?4.8 to 3.6 (Al) and 0.0 to 8.5 (Na) mmol_c kg^?1. The model predictions were good with values lying within 0.3 pH units (for the pH range 3.7 to 7.2), and 3% of CEC for individual cations. The most important proton buffer reaction seemed to be the dissolution of gibbsite and a large release of Al into the soil solution. When selectivity coefficients obtained by the modelling procedure were used to predict the field data for cation concentrations in the seepage water, they decreased in the following order: Na > K > Ca > Mg > Al. These calculated values were similar to the measured order: Na > Ca > K ≈ Mg > Al. Thus the options for managing these soils should be carefully chosen to avoid soil acidification which may result from inappropriate use of fertilizer during the cropping period.     

Variability of CO2 and N2O emissions during freeze‐thaw cycles: results of model experiments on undisturbed forest‐soil cores

The amounts of N₂O released in periods of alternate freezing and thawing depend on site and freezing conditions, and contribute considerably to the annual N₂O emissions. However, quantitative information on the N₂O emission level of forest soils in freeze‐thaw cycles is scarce, especially with regard to the direct and indirect effect of tree species and the duration of freezing. Our objectives were (i) to quantify the CO₂ and N₂O emissions of three soils under beech which differed in their texture, C and N contents, and humus types in freeze‐thaw cycles, and (ii) to study the effects of the tree species (beech (Fagus sylvatica L.) and spruce (Picea abies (L.) Karst.)) for silty soils from two adjacent sites and the duration of freezing (three and eleven days) on the emissions. Soils were adjusted to a matric potential of –0.5 kPa, and emissions were measured in 3‐hr intervals for 33 days. CO₂ emissions of all soils were similar in the two freeze‐thaw cycles, and followed the temperature course. In contrast, the N₂O emissions during thawing differed considerably. Large N₂O emissions were found on the loamy soil under beech (Loam‐beech) with a maximum N₂O emission of 1200 μg N m^–2 h^–1 and a cumulative emission of 0.15 g N m^–2 in the two thawing periods. However, the sandy soil under beech (Sand‐beech) emitted only 1 mg N₂O‐N m^–2 in the two thawing periods probably because of a low water‐filled pore space of 44 %. The N₂O emissions of the silty soil under beech (Silt‐beech) were small (9 mg N m^–2 in the two thawing periods) with a maximum emission of 150 μg N m^–2 h^–1 while insignificant N₂O emissions were found on the silty soil under spruce (0.2 mg N m^–2 in the two thawing periods). The cumulative N₂O emissions of the short freeze‐thaw cycles were 17 % (Sand‐beech) or 22 % (Loam‐beech, Silt‐beech) less than those of the long freeze‐thaw cycles, but the differences between the emissions of the two periods were not significant (P ≤ 0.05). The results of the study show that the amounts of N₂O emitted in freeze‐thaw cycles vary markedly among different forest soils and that the tree species influence the N₂O thawing emissions in forests considerably due to direct and indirect impacts on soil physical and chemical properties, soil structure, and properties of the humus layer.     

Differences in the chemical composition of dissolved organic matter from waste material of different sources

The chemical composition of waste-material-derived dissolved organic matter (DOM) was characterized by chemolytic analyses and ¹H, ¹³C and ³¹P nuclear magnetic resonance (NMR) spectroscopy. Dissolved organic matter was extracted by water from an aerobic fermented urban waste compost, a sewage sludge and a pig slurry and then fractionated using the XAD-8 method. The amount of water-extractable dissolved organic carbon (DOC) ranged from 3% in the sewage sludge to 22% in the pig slurry. Dissolved organic matter isolated from pig slurry was equally distributed between hydrophilic and hydrophobic DOC, whereas in the sewage-sludge-derived material the hydrophobic fraction was predominant. Dissolved organic C from the urban waste compost was mainly within the hydrophilic fraction. Wet-chemical analysis and ¹H- and ¹³C-NMR spectra showed that both DOM fractions from the urban waste compost were low in neutral, acidic and amino sugars as well as in lignin-derived compounds. In turn, the materials were rich in low-molecular-weight aliphatic compounds. The chemical structure of both fractions is probably the result of the intensive transformation of urban waste compost during its fermentation. The hydrophilic fractions of DOM from sewage sludge and pig slurry contained considerable amounts of carbohydrates but were also rich in low-molecular-weight aliphatics. The respective hydrophobic fractions had the largest contents of CuO-extractable phenols which may in part derive from sources other than lignin. By contrast with the other materials, the hydrophobic fraction from the pig slurry seemed to contain polymeric rather than low-molecular-weight material. The ³¹P-NMR spectrum of the hydrophilic DOM fraction from urban waste compost did not show signals of inorganic or organic P compounds while the spectrum of the hydrophobic fraction revealed traces of monoester P, diester P, and orthophosphate. ³¹P-NMR spectroscopy suggested that both the hydrophobic and hydrophilic fractions from pig slurry did not contain organic P. The hydrophilic DOM fraction from sewage sludge contained orthophosphate, organic monoester P and a little pyrophosphate. The hydrophobic fraction contained mainly organic diester P and smaller amounts of teichoic acids and organic monoester P. Considering that water-soluble fractions of urban waste compost contained no easily plant-available P and a low content of labile organics, we conclude that this material contains less labile nutrients and is more refractory than the soluble constituents of pig slurry and sewage sludge.     

Threadfin shad impacts phytoplankton and zooplankton community structures in channel catfish ponds

Plankton community structure and chlorophyll a concentration were compared in twelve 0.1 ha earthen ponds co‐stocked with channel catfish (Ictalurus punctatus Rafinesque, 1818) in a multiple‐batch culture (initial biomass=5458 kg ha^?1) and a planktivore, threadfin shad (Dorosoma petenense Güther, 1867; initial biomass=449 kg ha^?1), during the April–November growing season. We used a completely randomized design in a 2 × 2 factorial arrangement to test the planktivore level (presence or absence of threadfin shad) and channel catfish feeding frequency (daily or every third day). Channel catfish were fed a 32% protein feed to apparent satiation on days fed. The presence of threadfin shad affected phytoplankton and zooplankton community structure more than did feeding frequency, and the impact in ponds was more pronounced after 1 July. The numbers of all major groups of zooplankton were lower in ponds with threadfin shad, but were unaffected by the feeding frequency. Chlorophyll a concentration before 1 July was higher in ponds with threadfin shad and unaffected by the feeding frequency, whereas after 1 July it was higher in ponds without threadfin shad and that were fed daily. Phytoplankton community structure after 1 July was dominated by nuisance algal bloom genera of cyanobacteria in ponds without threadfin shad and by Bacillariophyceae in ponds with threadfin shad.     

Successful rearing of whiteleg shrimp Litopenaeus vannamei larvae fed a desiccation‐tolerant nematode to replace Artemia

The nematode Panagrolaimus sp. was tested as live feed to replace Artemia nauplii during first larval stages of whiteleg shrimp Litopenaeus vannamei. In Trial 1, shrimp larvae were fed one of four diets from Zoea 2 to Postlarva 1 (PL1): (A) Artemia nauplii, control treatment; (NC) nematodes enriched in docosahexaenoic acid (DHA) provided by the dinoflagellate Crypthecodinium cohnii; (N) non‐enriched nematodes; and (Algae) a mixture of microalgae supplemented in C. cohnii cells. In Trial 2, shrimp were fed (A), (NC) and a different treatment (NS) with nematodes enriched in polyunsaturated fatty acids (PUFAs) provided by the commercial product S.presso^®, until Postlarva 6 (PL6). Mysis 1 larvae fed nematodes of the three dietary treatments were 300 μm longer (3.2 ± 0.3 mm) than control larvae. At PL1, control shrimp were 300 μm longer (4.5 ± 0.3 mm) than those fed DHA‐enriched or PUFAs‐enriched nematodes. No differences were observed in length and survival at PL6 between control larvae and those fed DHA‐enriched nematodes (5.1 ± 0.5 mm; 33.1%–44.4%). Shrimp fed microalgae showed a delay in development at PL1. This work is the first demonstration of Panagrolaimus sp. suitability as a complete substitute for Artemia in rearing shrimp from Zoea 2 to PL6.     

Prediction of model pools for a long‐term experiment using near‐infrared spectroscopy

Fourty‐one soil samples from the “Eternal Rye” long‐term experiment in Halle, Germany, were used to test the usefulness of near‐infrared spectroscopy (NIRS) to differentiate between C derived from C₃ and C₄ plants by using the isotopic signature (δ¹³C) and to predict the pools considered in the Rothamsted Carbon (RothC) model, i.e., decomposable plant material, resistant plant material, microbial biomass, humified organic matter, and inert organic matter. All samples were scanned in the visible‐light and near‐infrared region (400–2500 nm). Cross‐validation equations were developed using the whole spectrum (first to third derivative) and a modified partial least‐square regression method. δ¹³C values and all pools of the RothC model were successfully predicted by NIRS as reflected by RSC values (ratio between standard deviation of the laboratory results and standard error of cross‐validation) ranging from 3.2 to 3.4. Correlations analysis indicated that organic C can be excluded as basis for the successful predictions by NIRS in most cases, i.e., 11 out of 16.