不同施肥对雷竹林土壤肥力及肥料利用率的影响

于2009年,在浙江省临安市的雷竹林中进行田间试验,以比较对照(不施肥)、常规施肥、缓释肥、竹笋专用肥和微生物肥对雷竹林土壤肥力质量、养分利用率和竹笋产量的影响。结果表明,雷竹林因施肥引起的土壤速效氮磷钾含量的过量积累和土壤酸化的程度按以下次序递减:常规施肥>专用复合肥≈缓释肥>微生物肥。常规施肥的肥料农学利用率最低,氮、磷和钾农学利用率分别为鲜笋4.7 kg kg-1、23.6kg kg-1和14.2 kg kg-1。与常规施肥处理相比,专用复合肥和缓释肥2个处理的氮、磷、钾农学利用率分别提高40.4%～53.2%、50.8%～90.7%和35.5%～39.5%,而微生物肥处理则分别提高38.9倍、67.8%和2.0倍。与对照相比,常规施肥处理的竹笋产量和经济效益分别提高了29.6%和14.3%,而专用复合肥、缓释肥、微生物肥3个处理的竹笋产量和经济效益分别提高了27.5%～29.0%和14.6%～17.8%。     

基于电磁感应仪数据的南疆棉田土壤电导率反演模型研究

土壤盐渍化是南疆棉田优质高产的主要障碍因素,定额灌溉可有效改良土壤盐渍化,EM38-MK2可快速、动态获取土壤盐渍化数据,适时监测土壤盐渍化,为棉田定额灌溉提供数字依据。针对电磁感应仪(EM38-MK2)的发射圈和接收圈设计了不同采样方案,在同一条田内采集了6个不同时期的土壤表观电导率数据及相应的剖面土样,分析了不同土壤采集方案及土壤含水量对表观电导率模型精度的影响,对比了以单一时期数据建模的局部模型和6个时期整体数据建模的全局模型的反演精度。结果表明,由单点采集电磁感应仪发射线圈位置土样所构建的模型精度更高,稳定性更好,能有效减少由采样带来的误差。当土壤含水量低于10%时,表观电导率与实测电导率之间的相关性较低,决定系数为0.58,反演模型只具备粗略估计实测电导率的能力;当土壤含水量高于10%时,表观电导率与实测电导率具有很好的相关性,决定系数达到0.80以上,反演模型具有较好的预测能力。EMH+EMV多测定模式下表观电导率与实测电导率之间的模型精度高于EMH或EMV单一测定模式,确立EC_h0.375+(EC_h0.75+EC_v0....     

西藏人工造林作业区土壤养分特征研究

为了掌握西藏拟造林作业区的土壤养分状况,对6个地市524个造林作业区的土壤样本进行了养分测定和分析。结果表明:(1)土壤全钾、有效磷、有机质和速效钾均处在"稍缺―极缺"水平:平均含量分别为6.26 g kg~(-1)、8.7 mg kg~(-1)、17.81 g kg~(-1)和83.6 mg kg~(-1),全氮和全磷处在"中等―稍缺"水平:平均含量分别为1.39 g kg~(-1)和0.72 g kg~(-1),碱解氮条件最好:平均含量为214.1 mg kg~(-1),全钾条件最差。(2)6地市除碱解氮外,土壤中分别存在2至5个养分元素的缺乏,多数作业区全钾、速效钾和有效磷不足。阿里地区最缺有机质,仅5.08 g kg~(-1),日喀则市、拉萨市和林芝市最缺全钾,仅5.19~6.42 g kg~(-1),山南市和昌都市最缺有效磷,仅3.9~5.4 mg kg~(-1)。此外,拉萨市、山南市和阿里地区多数作业区全氮不足,阿里地区、日喀则市和昌都市全磷不足,除昌都市外各地市有机质不足的作业区居多。(3)地市间土壤养分含量差异显著,但速效钾、有机质在日喀则市、拉萨市和山南市间差异均不显著。总体而言,西藏造林作业区土壤养分状况较差,通过对西藏造林作业区土壤养分的特征分析,可为作业区造林后树木生长状态的成因查寻提供科学依据,为进一步提高造林质量提供基础数据。     

Endomycorrhizae in a newly cultivated acidic meadow: Effects of three years of barley cropping,tillage, lime,and phosphorus on root colonization and soil infectivity

The dynamics of mycorrhizae under disturbance created by crop production is not well understood. A 3-year experiment was undertaken on a nutrient-poor and acidic land that had last been cultivated in the early 1970s. We observed the effects of cropping spring barley (Hordeum vulgare L.) under four P-fertilizer levels and four levels of lime, in a minimum (rototillage), a reduced (chisel), or a conventional tillage system, on the mycorrhizal receptiveness of the host (maximum level of mycorrhizal colonization, as measured at harvest) and soil infectivity most probable number method. The host receptiveness decreased with time, while crop yields and soil infectivity increased simultaneously with time. Liming increased mycorrhizal colonization of barley roots and soil infectivity. P additions decreased root colonization but did not significantly affect the most probable number valuse. Slightly higher soil infectivity estimates were found under reduced tillage.     

Soil moisture pre-treatment effects on enzyme activities as indicators of heavy metal-contaminated and reclaimed soils

Heavy metal contamination can inhibit soil functions but it is often difficult to determine the degree of pollution or when soil reclamation is complete. Enzyme assays offer potential as indicators of biological functioning of soils. However, antecedent water content of soil samples may affect the outcome of biological measurements. In Mediterranean regions, for much of the year ‘field moist’ surface soil can have water content similar to that of air-dry samples. The objectives of this study were to: (1) determine the sensitivity of a range of enzyme assays to detect the degree of pollution from a heavy metal mine spill; (2) evaluate rewetting field-dry soil as a pre-treatment for enzyme assays; and (3) test multivariate analysis for improving discrimination between polluted, reclaimed and non-polluted soils. The Aznalcóllar mining effluent spill provided a unique opportunity to address these objectives. This accident released toxic, heavy metal-contaminated (As, Bi, Cd, Cu, Pb, Tl, Zn…) and acid tailings into the Guadiamar watershed (SW Spain) in 1998, severely affecting the riparian zone along more than 4000 ha. Contaminated soils were collected from the highly polluted upper watershed and less polluted lower watershed along with reclaimed soil at both sites. Enzyme activities (phosphatases, arylsulfatase, β-glucosidase, urease and dehydrogenase) were assessed on both field-moist samples and soils rewetted to 80% of water-holding capacity and then incubated at 21 °C for 7 d prior to the assay. The reclaimed soils had higher activities than polluted soils but, typically, 1.5-3 times lower levels of activity than the non-polluted soil. Regardless of the moisture pre-treatment, all enzymes showed significant effects due to pollution, with urease and β-glucosidase showing the greatest discrimination between degrees of contamination. In general, rewetting field-dried soils increased activities on non-polluted and reclaimed soils which improved discrimination with polluted soils. Another method to increase the potential of soil enzyme activities to detect soil contamination could be to combine them in multivariate analysis, which provides a more holistic representation of the biochemical and microbial functionality of a soil.     

拱棚辣椒——大球盖菇轮作改良土壤试验

为解决拱棚蔬菜种植连作障碍问题,进行了辣椒—大球盖菇轮作试验。结果表明:通过轮作,土壤有机质含量增加0.41%,pH值提高0.46,容重减少0.06 g/cm3,化肥投入量减少约30%,农药使用量降低约50%;辣椒根腐病发病率减少7.5%,667 m~2产量增加523 kg。综合来看,辣椒与大球盖菇轮作可改良土壤,防控土传病害,提高产量,有利于绿色食品的生产。     

Impact of plant cell wall network on biodegradation in soil: Role of lignin composition and phenolic acids in roots from 16 maize genotypes

Residue quality is a key factor governing biodegradation and the fate of C in soil. Most investigations of relationships existing between crop residue quality and soil decomposition have been based on determining the relative proportions of soluble, cellulose, hemicellulose and lignin components. However, cell wall cohesion is increased by tight interconnections between polysaccharides and lignin that involve cross-linking agents (phenolic acids). The aim of this study was to determine the role of lignin composition and phenolic acids on short- to medium-term decomposition of maize roots in soil. Sixteen maize genotypes, presenting a range of chemical characteristics related to root lignin and phenolic acids, were used. The main components were characterized by Van Soest (VS) extraction and cell wall acid hydrolysis, and the non-condensed Syringyl and Guaicyl lignin monomers, esterified phenolic acids and etherified phenolic acids were determined. Maize roots were then incubated in soil under controlled conditions (15 °C, −80 kPa moisture) for 796 days. Results showed that VS extraction over-estimated the structural hemicellulose content and that VS lignin was more recalcitrant than Klason lignin. The tremendous effect of cell wall chemical characteristics was shown by marked variations (almost two-fold differences in C mineralization), between the 16 maize roots. Decomposition was controlled by soluble residue components in the short term whereas lignin and the interconnections between cell wall polymers were important in the long-term. Notably the cell wall domain rich in non-condensed lignin and esterified phenolic acids was prone to decomposition whereas the presence of etherified ferulic acids seemed to hamper cell wall decomposition.     

A soil chronosequence in the semi-arid environment of Patagonia (Argentina)

Soil development with time was investigated on beach ridges with ages ranging from about 1380 to 6240 ¹⁴C-years BP at the eastern coast of central Patagonia. The main pedogenic processes are accumulation of organic matter and carbonate leaching and accumulation within the upper part of the soils. Soil formation is strongly influenced by incorporation of eolian sediments into the interstitial spaces between the gravel of which the beach ridges are composed. Different amounts of eolian material in the soils lead to differentiation into Leptosols (containing ≤ 10% fine earth in the upper 75 cm) and Regosols (containing > 10% fine earth). Soil depth functions and chronofunctions of organic carbon, calcium carbonate, pH, Ca:Zr, Mg:Zr, K:Zr, Na:Zr, Fe:Zr, Mn:Zr, and Si:Al (obtained from X-ray fluorescence analysis) were evaluated. To establish soil chronofunctions mean values of the horizon data of 0–10 cm below the desert pavement were used, which were weighted according to the horizon thicknesses. The depth function of pH shows a decrease towards the surface, indicating leaching of bases from the upper centimeters. Chronofunctions of pH show that within 6000 radiocarbon years of soil development pH drops from 7.0 to 6.6 in the Leptosols and from 8.1 to 7.5 in the Regosols. The higher pH of the Regosols is due to input of additional bases from the eolian sediments. Chronofunctions of Ca:Zr and K:Zr indicate progressive leaching of Ca and K in the Regosols, showing close relationships to time (R² = 0.972 and 0.995). Na leaching as indicated by decreasing Na:Zr ratios shows a strong correlation to time only in the Leptosols (R² = 0.999). Both, Leptosols and Regosols show close relationships to time for Fe:Zr (R² = 0.817 and 0.824), Mn:Zr (R² = 0.940 and 0.803), and Si:Al (0.971 and 0.977), indicating enrichment of Fe and Mn and leaching of Si. Leaching of mobile elements takes place on a higher level in the Regosols than in the Leptosols from the beginning of soil formation. Hence, a significant part of the eolian sediments must have been incorporated into the beach ridges very soon after their formation.     

Soil nitrogen mineralization not affected by grass species traits

Species N use traits was evaluated as a mechanism whereby Bromus inermis (Bromus), an established invasive, might alter soil N supply in a Northern mixed-grass prairie. We compared soils under stands of Bromus with those from three representative native grasses of different litter C/N: Andropogon gerardii (Andropogon), Nassella viridula (Nassella) and Pascopyrum smithii (Pascopyrum); in ascending order of litter quality. Net mineralization (per g soil N) measured in year-long laboratory incubations showed no differences in comparisons of Bromus with two of the three native grasses: Andropogon and Nassella. Higher mineralization in Pascopyrum stands relative to Bromus was consistent with its higher litter quality. However, an unusually high occurrence of an N-fixing legume in Pascopyrum stands, potentially favoring high mineralization rates, confounded any conclusions regarding the effects of plant N use on N mineralization. Instead of an initial flush of net mineralization, as would be expected in laboratory incubation, we observed an initial lag phase. This lag in net N mineralization coincided with high microbial activity (respiration) that suggests strong N limitation of the microbial biomass. Further support for the importance of immobilization initially came from modeling mineralization dynamics, which was explained better when we accounted for microbial growth in our model. The absence of strong differences in net mineralization beneath these grasses suggests that differences in plant N use alone were unlikely to influence soil N mineralization through substrate quality, particularly under strong N control of the microbial biomass.     

An inter-laboratory comparison of multi-enzyme and multiple substrate-induced respiration assays to assess method consistency in soil monitoring

The use of indicators in soil monitoring schemes to detect changes in soil quality is receiving increased attention, particularly the application of soil biological methods. However, to date, the ability to compare information from different laboratories applying soil microbiological techniques in broad-scale monitoring has rarely been taken into account. This study aimed to assess the consistency and repeatability of two techniques that are being evaluated for use as microbiological indicators of soil quality: multi-enzyme activity assay and multiple substrate-induced respiration (MSIR). Data were tested for intrinsic (within-assay plate) variation, inter-laboratory repeatability (geometric mean regression and correlation coefficient) and land-use discrimination (principal components analysis). Intrinsic variation was large for both assays suggesting that high replicate numbers are required. Inter-laboratory repeatability showed diverging patterns for the enzyme assay and MSIR. Discrimination of soils was significant for both techniques with relatively consistent patterns; however, combined laboratory discrimination analyses for each technique showed inconsistent correspondence between the laboratories. These issues could be addressed through the adoption of reliable analytical standards for biological methods along with adequate replication. However, until the former is addressed, dispersed analyses are not currently advisable for monitoring schemes.