湿地植物的移栽是人工湿地修复的主要组成部分，移栽期间，植物从超苗、运输到栽植时间的长短等都直接影响到湿地植物移栽后的成活率和生长状况。以南四湖人工湿地中芦竹（Arundo donax）、芦苇（Phragmites communis）和香蒲（Typha latifolia）为研究对象，研究了移栽胁迫在超苗后24h内对3种植物的影响。结果表明：起苗后，芦苇的生物量和叶绿素含量下降速率最小，芦竹的叶绿素含量下降最快；叶绿素荧光方面，芦竹光合系统Ⅱ（PSⅡ）的潜在活性（Fv/Fo）和潜在最大光合能力（Y）最大，光保护能力（NPQ）也最强；香蒲的实际光合效率（φPSⅡ）最大；3种湿地植物的光化学淬灭（qP）、光保护能力和实际光合效率在5h左右即降至最低，而潜在最大光合能力下降至22h后达到稳定的较低值；芦苇的φPSⅡ、Y、qP、NPQ和香蒲的φPSⅡ、qP均与叶绿素含公交日减少呈极显著的正相关关系（P<0.01）。以上结果表明，移栽胁迫对3种湿地植物产生了严重的影响，且影响程度为芦苇>香蒲>芦竹，起苗后栽植时间分别不宜超过4-6h、6-8h和14-16h，以保证移栽后的植物成少多样性和正常生长。

为了探索西南喀斯特地区土壤有效磷的变异规律，采用野外采样与实验室分析相结合的方法，对贵州省荔波县及普定且不同石漠化阶段典型土壤有效磷含量进行了研究。结果表明：①非石漠化土壤与轻度石漠化土壤有效磷质量分数呈显著差异，非石漠化的黑色石灰土有效磷质量分数高达9.29×10-6，随着石漠化程度的加深，土壤有效磷含量逐渐减少；②土壤剖面中有效磷质量分数均表现为表土层高于心土层，且随着石漠化的发展，有效磷质量分数在剖面中减少得越慢；③土壤有效磷质量分数随时间有显著变化，有效磷从5月开始积累，7月后出现下降趋势，9月或11月又逐渐升高，到翌年1月达到峰值；④土壤有效磷与有机质和黏粒质量分数均呈极显著线性相关，相关系数分别为0.9017和-0.7772。这说明增加有机质的积累可提高土壤有效磷质量分数，对防治石漠化具有重要意义。

我国西南喀斯特地区的石漠化问题已经严重制约了当地社会经济发展，揭示土壤养分变化规律对该区域生态系统适应性修复具有重要意义。以贵州省荔波县和普定县为研究区，通过野外采样和实验室分析，对西南喀斯特地区典型石漠化阶段土壤的铵态氮、硝态氮变异状况进行了研究。结果表明：（1）非石漠化的黑色石灰土中铵态氮、硝态氮的含量分别高达11.61和38.01mg·kg-1。随着石漠化程度的加深，土壤中氮素含量逐渐减少。统计表明非石漠化土壤与轻度石漠化土壤铵态氮含量呈显著差异，硝态氮含量呈极显著差异。（2）土壤铵态氮含量表现为表土层高于心土层，而硝态氮在剖面中分布规律不一致。（3）土壤氮素含量随时间有明显变化。铵态氮和硝态氮含量均在7月达到峰值，然后呈现出下降趋势，至11月或翌年1月又逐渐增加。（4）土壤有机质与铵态氮和硝态氮含量呈一元二次线性相关，相关系数分别为0.7671和0.9493，均达到极显著水平；铵态氮和硝态氮含量也呈极显著的线性相关，相关系数为0.7743。喀斯特地区通过封山育林等措施增加有机质积累可提高土壤氮素含量，对防治石漠化具有重要意义。

酸雨中的SO2-4和NO-3等阴离子在土壤中迁移时会引起大量的盐基离子淋溶，导致土壤退化。用室内土柱模拟SO2-4和NO-3在红壤旱地各土层中的垂直穿透状况，并用Hydrus-1D模型对试验结果进行了拟合和预测。结果表明，NO-3在红壤各层中的穿透速度较快，其中在耕作层的穿透曲线峰值最高，C/C0达到0.39，峰值高低顺序依次为：耕作层>母质层>淋溶层>犁底层。SO2-4在土柱中的穿透速度远低于NO-3，穿透曲线有明显的拖尾现象。其在各土层的穿透时间依次为：母质层>犁底层>淋溶层>耕作层；而其峰值高低顺序依次为：耕作层>淋溶层>母质层>犁底层，最高点耕作层的顶点C/C0 仅为0.22。用Hydrus-1D模型对试验结果进行模拟，所得的SO2-4和NO-3穿透土壤的浓度模拟值与其实测值均呈极显著的正相关关系。利用数学模拟获得了饱和导水率和垂直扩散率等溶质运移参数，并预测了研究区酸雨后SO2-4和NO-3在红壤耕层的迁移状况，表明SO2-4会在酸雨结束后持续淋溶，从而影响土壤中Ca、Mg等盐基离子的淋失。

The influence of texture, bulk density, and organic matter content on the process of nitrate vertical transport in the three main paddy soils (Bai soil, Huangni soil and Wushan soil) of the Tai Lake region were studied in the soil columns. Breakthrough curves (BTC) were obtained separately for each of thirteen horizons. The results were as follows: vertical transport velocity of nitrate decreased, and the BTCs of nitrate were more dispersed, in each horizon from the surface layer to the bottom in every soil profile. Among the three soils, the average flux ofWushan soil was the least and its nitrate BTC was the most dispersed. Under saturated conditions, nitrate penetrated the soil column quickly. The transport of nitrate was affected by clay content. As the clay content increased, nitrate outflow was retarded, and the peak concentration was reduced. Nitrate BTCs rose and fell gently when the nitrate concentration was lower. All nitrate BTCs were asymmetric, and tailing was more obvious when clay content was high. Soil bulk density and the organic matter content also affected the vertical transport of nitrate. Low bulk density and high organic matter content were each associated with faster nitrate transport. An analytical solute transport model (CXTFIT 2.0) was used to estimate the nitrate dispersion coefficient and average pore-water velocity from the observed breakthrough curves. The results showed that the analytical solute transport model was suitable in fitting the observed nitrate transport in these soils. The dispersion coefficient was found to be a function of average pore-water velocity.