内蒙古高原紫花苜蓿土壤氮素丰缺指标和推荐施氮量

为了给内蒙古高原紫花苜蓿（Medicago sativa L.）测土施氮奠定科学基础，本研究采用“零散实验数据整合法”和“养分平衡-地力差减法”新应用公式，开展了该自然区域紫花苜蓿土壤氮素丰缺指标和推荐施氮量研究。结果表明：内蒙古高原生长第1年紫花苜蓿土壤碱解氮第1~6级丰缺指标为≥48，20~48，8~20，4~8，2~4和<2 mg·kg^-1，土壤全氮第1~5级丰缺指标为≥1.4，0.8~1.4，0.4~0.8，0.2~0.4和<0.2 g·kg^-1，土壤有机质第1~6级丰缺指标为≥17，10~17，6~10，3~6，2~3和<2 g·kg^-1。当紫花苜蓿目标产量9~18 t·hm^-2、氮肥利用率40%时，内蒙古高原紫花苜蓿第1~6级土壤推荐施氮量分别为0，68~135，135~270，203~405，270~540和338~675 kg·hm^-2。     

面向养殖水体分布差异的COD光谱法检测研究

面向养殖水体，传统光谱法对化学需氧量（Chemical Oxygen Demand，COD）检测模型构建的基础：源域（现有样本库）与目标域（检测地水体）间光谱数据独立同分布。但是当源域与目标域分布间存在差异时，由源域得到的低误差模型常在目标域上表现下滑。针对该问题，提出面向UV Vis光谱的域对抗训练网络（DAUVwpNet），将分布不同的源域和目标域数据映射至相同分布的特征空间中，使其在该空间的分布距离尽可能接近，从而在特征空间中对源域训练的目标函数也可以迁移至目标域上，以降低模型在目标域的误差。试验表明：面向同一批测试数据，DAUVwpNet的预测误差为0.78，要低于传统模型的预测误差（0.85）；DAUVwpNet预测值与实测值间相关系数为0.95，要高于传统模型的相关系数（0.89）。表明了该网络能够较好对齐两域特征空间数据分布，降低因分布差异带来的COD检测误差。     

Soil protection for a sustainable future: options for a soil monitoring network for Ireland

The increased recognition of the importance of soil is reflected in the UN Post‐2015 Development Agenda with sustainable development goals that directly and indirectly relate to soil quality and protection. Despite a lack of legally binding legislation for soil protection, the European Commission remains committed to the objective of soil protection. However, the achievement of a legally binding framework for soil protection relies on the implementation of a soil monitoring network (SMN) that can detect changes to soil quality over time. As beneficiaries do not pay for the provision of soil information, the options for soil monitoring are limited. The use of existing data sets should be considered first. Using Ireland as an example, this research explored the opportunities for a SMN for Ireland considering three existing national data sets. The options for a SMN are considered in terms of their spatial and stratified distribution, the parameters to be measured and an economic analysis of the options proposed. This research finds that for Ireland, either a 10 or a 16 km² grid interval stratified by land use and drainage class offers the best potential in relation to the spatial distribution of existing data sets to reflect local data at a national level. With existing data, the stratified SIS data using the 16 km² grid offers the best value for money, with baseline costs for analysis, excluding field costs, of between €706 481 and €2.8 million. Acknowledging the impossibility of measuring all parameters with ideal frequency, this study proposes a two‐tier system for optimized monitoring frequency. Parameters must anticipate future policy requirements. Finally, the implementation of a SMN must be accompanied by standardized methods, defined thresholds and action mandates to maintain soil quality within allowable limits.     

Factors influencing decline in soil pH in Hawaiian Eucalyptus and Albizia plantations

Soil pH declined from 5.9 to 5.0 in 8 years beneath plantations of Eucalyptus saligna (Sm.) in Hawaii. In stands of Albizia falcataria, (L.) Fosberg, the soil pH change was more dramatic, declining from 5.9 to 4.6. We measured several components of soil acidity beneath four mixtures of the two tree species to gain insight on the processes responsible for the decline in soil pH. These components were studied using an empirical method of comparing acid quantity, degree of neutralization (depletion of base cations), and acid strength. The decline in soil pH differed between species as a result of differences in the degree of neutralization of the soil exchange complex; the larger decrease in soil pH under Albizia was produced by greater acidification of the exchange complex. Empirical titration curves suggested that differences in acid strength moderated the divergence in soil pH beneath the species. Had the acids accumulating in the soil under Albizia been as strong as those in the Eucalyptus soil, the difference in soil pH would have been greater. Though the two species had contrasting effects on soil pH, the differences in degree of neutralization, responsible for the pH decline, were small compared with differences in the amount of cations stored in tree biomass. Continued supply of nutrient cations (from weathering or fertilization) will ultimately control both the extent of soil pH decline and the level of productivity sustained by the forest.     

Effects of 'Agri-SC' soil conditioner on soil structure and erodibility: some further observations

Abstract. In field and laboratory experiments the conditioner‘Agri-SC’has shown improvements in the structure of loamy sand soils in east Shropshire, UK. It resulted in statistically significant decreases in soil bulk density values and increases in soil porosity and aggregate stability. Further experiments are in progress on both loamy sand and silt loam soils.     

越桔对长白山区酸性土壤的适应性

在长白山区３种类型的酸性土壤上栽培越桔，研究结果表明，｀北空＇、｀美登＇、｀斯卫克＇在暗棕色森林土上生长结果较好，树体营养关系合理；在草甸沼泽土上３个种生长结果均较差，｀北空＇在水湿地潜育土上生长旺盛，但产量低，综合分析认为，矮丛越桔类型品种对我国长白山区酸性土壤条件的适应性较强，半高丛越桔则适应性差，针对长白山不同类型的土壤，以美登最为适宜暗棕色森林土栽培，斯卫克次之；水湿地潜育土可栽培美登品种     

海南半干旱地区芒果间作柱花草及作物效益初探

1998～2001年在海南东方干旱半干旱燥红土芒果与柱花草及作物间作示范区的实践表明，在芒果园间作扁豆、柱花草、番薯和花生可显著提高芒果挂果率，芒果收入比芒果单种分别增加97．79％、98．53％、54．41％和48．28％，果园综合收入比芒果单种分别增加108．55％、98．53％、97．98％和91．58％，同时，果园间作可显著提高果园土壤有机质、全氮、速效磷的含量和土壤pH值，间作扁豆使土壤有机质含量增加53．3％，间作柱花草使土壤全氮增加43．92％，速效磷增加78．16％，从而达到了以短养长、增加农民收入、提高土壤肥力和持续管理利用的目的。     

南、北盘江流域（贵州部分）土地退化类型及机理

在对珠江上游南、北盘江流域（贵州部分）调查的基础上，分析了该流域土地退化成因，原生生态系统的脆弱性只是导致土地退化的一个方面。而人为活动才是土地退化的主要因素。根据土地退化的成因、结果，将珠江上游南、北盘江流域（贵州部分）土地退化划分为：土壤侵蚀、石漠化、土地贫瘠、土壤污染、土壤破坏五种类型，并提出了退化土地的管理及恢复措施。     

小麦深施肥播种机应用与效果初探

通过新型的小麦深施肥播种机的应用，实现了小麦深施肥播种，使种、肥分箱不同床，解决了过去化肥对种芽、幼苗的危害问题。达到了节本增效、增产增收的目的。     

振动深松鼠道犁的试验分析

石河子及周围垦区的土壤粘重且呈盐碱性,由于犁底层的存在,土质很难得到改善。而深松作业则是打破犁底层的主要农业技术措施之一。为此,论述了IZSS-600-3型振动深松鼠道犁的构造和工作原理,分析了试验因素对试验指标的影响,为机具今后在农业生产中的应用和推广提供了依据和参考。