蓝藻水华预报模型及基于遗传算法的参数优化

蓝藻水华预报是应对水危机,保障水资源供给的一项重要工作。以太湖北部三湾(竺山湖、梅梁湾、贡湖)为研究对象,采用动态空间环境建模技术,构建了蓝藻水华预报模型,并通过实地观测建立了模拟的初始参数集。利用2008年04-09月太湖水环境、气象等实测数据,采用遗传算法优化叶绿素a浓度预报模型中敏感度较高的4个参数。研究结果表明,该模型在蓝藻水华空间分布的预报上达到了一定的精度;采用遗传算法能全面、高效地进行参数优化,降低了模拟结果的相对残差,提高了模型预报精度。     

基于VGPM模型和MODIS数据估算梅梁湾浮游植物初级生产力

基于MODIS影像数据反演的2009年2月份至12月份太湖梅梁湾水域表面叶绿素a、悬浮物浓度以及水温数据,结合初级生产力垂向归纳模型(Vertically Generalized Production Model:VGPM)估算获得梅梁湾2009年逐月平均日初级生产力时空分布。结果表明,梅梁湾2009年年平均日初级生产力及逐月平均日初级生产力空间分布差异显著,呈现从湾内向湾口逐渐递减的趋势。时间序列分析显示,梅梁湾初级生产力季节差异显著,夏季>秋季>春季>冬季,全年初级生产力主要集中在夏季,占47.4%。通过分析VGPM模型中几个输入参数对初级生产力的影响,发现悬浮物浓度与标准化初级生产力存在显著负冥函数关系,反映沉积物再悬浮引起的悬浮物浓度增加能降低水体初级生产力。温度对初级生产力也有一定的调控与制约,与初级生产力呈现正相关趋势,在低于21℃的温度范围内与最大光合作用速率呈现正相关。     

基于模型数据融合的千烟洲亚热带人工林碳水通量模拟

人工林生态系统是我国森林生态系统的重要组成部分,在全球碳平衡中的作用越来越受到重视.利用千烟洲亚热带人工针叶林通量观测站的碳水通量和气象观测数据,通过模型数据融合方法对碳水循环过程模型——SIPNET模型关键参数进行反演,模拟了2004-2009年千烟洲人工林生态系统的碳水通量.结果表明:仅用碳通量观测数据优化模型参数时,净生态系统碳交换量(NEE)模拟效果较好(R2=0.934),而生态系统蒸散(ET)模拟效果较差(R2=0.188);同时用碳水通量观测数据优化时,NEE模拟效果稍差(R2=0.929),但ET模拟效果显著提升(R2=0.824),说明利用碳水通量观测数据同时优化,SIPNET模型才能较好地模拟试验站点碳水通量.在此基础上,开展了人工林生态系统碳通量对降水变化响应的敏感性分析,发现降水量减少对光合作用的影响比对呼吸作用的影响更为强烈,且碳水通量同时参与优化时模型才能较好地模拟碳通量随降水减少而快速降低的趋势,表明如果不能同时利用碳水通量进行参数优化,模型无法正确揭示生态系统碳循环对降水变异的响应.     

湖泊生态系统动力学模型研究进展

从系统分析在湖泊生态系统动力学研究中的作用出发,对湖泊生态系统的动力学建模过程、方法和软件等进行了总结.在此基础上,综述了国内外湖泊生态系统动力学模型的发展.从1960年代至今,湖泊生态系统动力学模型从简单的零维模型发展到复杂的水质水动力学生态综合模型和生态结构动力学模型,如LakeWeb模型.中国的湖泊生态系统动力学模型研究始于20世纪80年代,主要集中在滇池、太湖、东湖和巢湖等富营养化严重的湖泊以及其他水体.目前,已经开发一些软件用于湖泊生态系统动力学模拟,主要有CEQUALICM、WASP、AQUATOX、PAMOLARE、CAEDYM等,以及用来模拟湖泊能流的软件ECOPATH.湖泊生态系统动力学模型还在监测、数据共享和模型结构、参数选取和不确定性分析等方面存在不足,需在今后的研究中加以改进.     

太湖沉积物中氮循环菌的微生态

采用荧光原位杂交(FISH)技术,对梅梁湾和贡湖湾湖区沉积物中微生物的主要种群,及氮循环功能微生物的数量和分布进行了研究,发现随着深度增加,细菌和古菌数量均逐渐减少,但古菌在总菌数中所占比例有所增加;梅梁湾沉积物中氨氧化细菌及亚硝酸氧化细菌的数量均高于贡湖湾,随深度增加,两者数量均逐渐减少,但在贡湖湾其占总菌数的比例高于梅梁湾,表明水生植物可能对沉积物中微生物组成及氮循环有重要影响。泉古菌在太湖沉积物中普遍存在且数量高于氨氧化细菌,说明其在淡水湖泊中对氮循环可能有重要作用。     

内陆湖泊水体固有光学特性的典型季节差异

固有光学特性是水体光学性质的重要内容,是水色反演分析模型建立的基础.本研究利用定量滤膜技术（QFT）和后向散射测量仪BB9,对太湖梅梁湾夏、冬季水体的有色可溶性有机物(CDOM)吸收系数、总悬浮物吸收系数和总悬浮物后向散射系数进行了观测.在分别对两季节水体组分吸收系数、后向散射系数光谱特征分析的基础上,阐明其季节差异性,并结合水质参数的变化,揭示导致两季节水体固有光学特性不同的原因,达到通过固有光学量反映水环境状态的目的.初步建立了后向散射系数与悬浮物浓度的关系模型,为分析模型的构建提供了参数保障.     

不同样本方案下遗传元胞自动机的土地利用模拟及景观评价

利用元胞自动机(cellular automata,CA)模拟土地利用情景,有助于理解其变化机理,并为土地资源持续利用提供空间决策支持.本文基于生物进化过程的遗传算法(genetic algo-rithm,GA)将CA参数编码成为染色体,在模拟结果与真实结果差异值的引导下,通过选择、杂交和变异算子使最优的染色体得以遗传和保留,从而建立智能优化的元胞自动机模型.以浙江省嘉兴市1992-2008年土地利用变化为例,分别利用6%(66个·km-2)和3%(33个·km-2)两种样本方案构建遗传CA模型进行土地利用变化模拟,并通过混淆矩阵、Kappa系数和景观指数对模拟结果进行评估.结果表明:遗传CA模拟结果能在数量、位置和景观格局上以超过80%的水平接近真实分类,且大样本量构建的遗传CA的模拟精度更高;2008年的模拟精度和景观综合指数低于2001年,表明遗传CA的模拟精度和景观综合指数随模拟时间而衰减.     

利用模拟退火算法优化Biome-BGC模型参数

生态过程模型建立在明确的机理之上,能够较好地模拟陆地生态系统的行为和特征,但模型众多的参数,成为模型具体应用的瓶颈。本文以Biome-BGC模型为例,采用模拟退火算法,对其生理、生态参数进行优化。在优化过程中,先对待优化参数进行了选择,然后采取逐步优化的方法进行优化。结果表明,使用优化后的参数,模型模拟结果与实际观测更为接近,参数优化能有效地降低模型模拟的不确定性。文中参数优化的过程和方法,可为生态模型的参数识别和优化提供一种实例和思路,有助于生态模型应用区域的扩展。     

植被光合呼吸模型在千烟洲亚热带常绿针叶林的优化及验证

碳循环模型参数的确定和优化对生态系统净CO₂交换(NEE)的模型计算至关重要。该文利用2010-2012年ChinaFLUX千烟洲站点的通量观测资料, 对植被光合呼吸模型(VPRM)的参数进行了优化。通过比较两种不同的拟合方案, 发现利用传统光响应方程得到的参数不适用于VPRM, 而利用模型自身反演方案拟合得到的参数最大光量子效率(λ)达0.203, 大于C₃植物平均值, 但与其他相关研究结果吻合。采用VPRM模型反演方案优化得到的参数后, VPRM能较准确地模拟千烟洲站不同季节的NEE。其对全年半小时NEE模拟的平均误差为-0.86 μmol·m^-2·s^-1, 相关系数为0.72。模型可准确地模拟生长旺季NEE平均日变化, 但低估了非生长旺季白天吸收峰值约52%。通过个例分析发现, VPRM模型可以准确模拟晴天条件下NEE的时间变化, 但对阴雨天条件下NEE的模拟还存在较大的不确定性。该研究将有助于进一步改进CO₂通量及浓度的区域数值模拟。     

基于FLUXNET观测数据与VPM模型的森林生态系统光合作用关键参数优化及验证

生态系统总初级生产力(GPP)是全球生态系统碳循环研究的重要组成部分.植被最大光能利用率(ε_max)是陆地生态系统GPP模拟的关键参数.本文基于植被光合模型(VPM)和全球通量网(FLUXNET)40个站点(179条站点年数据)的涡度相关通量观测数据,采用单因素轮换法对VPM模型进行参数敏感性分析,并利用交叉验证法对全球森林生态系统的光合作用关键参数进行优化和验证.结果表明:森林生态系统GPP模型受ε_max、光合最高温度(T_max)以及光合最适温度(T_opt)的影响最大;优化后的ε_max在不同植被类型之间存在明显差异,介于0.05～0.08 μmol CO₂·μmol^-1 PAR,常绿阔叶林>常绿针叶林>混交林>落叶阔叶林;优化后的森林生态系统T_max为38～48 ℃,T_opt为18～22 ℃;利用分植被类型优化后的模型关键参数,VPM模型可较好地模拟全球主要森林生态系统GPP的季节和年际变化.     

Grazing on filamentous algae by herbivorous zooplankton

SUMMARY. 1. Feeding experiments were conducted to examine filtering rates and selectivity of a variety of zooplankton taxa (including cladocerans, copepods and a rotifer) for filamentous diatoms, green and blue-green algae.
2. Most herbivores were capable of consuming some filamentous algae at rates similar to or higher than those on unicellular algae. Only feeding of Diaphanosoma brachyurum Liéven and Moina micrura Kurz seemed to be primarily limited by the filamentous morphology.
3. Filtering rates and selectivities of most herbivores were much higher for the diatom Melosira granulata angustissima Müller than for similarly sized blue-green algal filaments, indicating that chemical factors strongly influence consumption of filamentous algae.
4. The toxic blue-green algal filament Anabaena flos-aquae De Brébisson NRC 44–1 had a much strong inhibitory effect on zooplankton feeding than other filaments. The only herbivores that were not inhibited by this strain have been shown to be resistant to blue-green algal toxins, or strongly avoided consuming the blue-green alga. These results indicate that the inhibitory effect of filamentous algae is due more to toxic or noxious chemicals than to the filamentous morphology.
5. Selectivities of zooplankton for filamentous algae were largely independent of herbivore body size. The small-bodied cladoceran Bosmina longirostris Müller had the highest selectivities for filamentous algae.     

Process-based simulation library SALMO-OO for lake ecosystems. Part 1: Object-oriented implementation and validation

The process-based simulation library SALMO-OO represents an object-oriented implementation of mass balance equations for pelagic food webs consisting of diatoms, green algae, blue-green algae and cladocerans, as well as nutrient cycles based on PO₄–P, NO₃–N, DO and detritus in lakes. It gains its structural flexibility from alternative process representations for algal growth and grazing, and zooplankton growth and mortality.Generic model structures of SALMO-OO were determined for three lake categories classified by circulation types reflecting the local climate and morphometry of lakes, and by trophic states reflecting community structures and habitat properties of lakes.Data of six lakes belonging to either of the three lake categories were used to validate the lake category specific structures of SALMO-OO.The study has demonstrated that: (1) JAVA suits well for object-oriented implementations of ecological simulation libraries, (2) object-oriented simulation libraries facilitate the identification of generic model structures for ecosystem categories.     

Regulation of blue-green algal buoyancy and bloom formation by light,inorganic nitrogen,C02, and trophic level interactions

In highly eutrophic ponds, buoyancy of the gas-vacuolate blue-green alga Anabaenopsis Elenkinii (Miller) was regulated by complex interactions between chemical and physical parameters, as well as by biological interactions between various trophic levels. Algal buoyancy and surface bloom formation were enhanced markedly by decreased light intensity, and to a lesser extent by decreased CO₂ availability and increased availability of inorganic nitrogen. In the absence of dense populations of large-bodied Cladocera, early season blooms of diatoms and green algae reduced light availability in the ponds thus creating conditions favorable for increased buoyancy and bloom formation by A. Elenkinii. The appearance of blue-green algal blooms could be prevented by a reduced density of planktivorous fish, which allowed development of dense cladoceran populations. The cladocerans limited the growth of precursory blooms of diatoms and green algae, and given the resulting clear-water conditions, buoyancy of A. Elenkinii was reduced, and blue-green algal blooms never appeared.     

Influence on phytoplankton biomass in lakes of different trophy by phosphorus in lake water and its regeneration by zooplankton

In 49 unpolluted lakes of north-eastern Poland the biomass of algae in summer is significantly related to the concentration of total phosphorus and to the rate of phosphorus regeneration by zooplankton. Using a model with equations describing these relationships, the biomass of blue-green algae and other phytoplankton groups was predicted for 14 polluted lakes. A good approximation of actual values was obtained only for the biomass of blue-green algae calculated from the estimated rate of P regeneration by zooplankton in these lakes. It is hypothesized that more-or-less edible algae of other classes did not show dependence on the rate of input of regenerated P because their biomass was heavily reduced by grazing of zooplankton.     

Alga consumption of four dominant planktonic crustaceans in Lake Balaton (Hungary)

Between 1981–83 the gut contents ofDaphnia galeata, D. cucullata, Eudiaptomus gracilis, andCyclops vicinus were examined with light and scanning electron microscope to obtain information on the feeding of these species in Lake Balaton. The twoDaphnia species feed mainly on abioseston, and it is assumed that their primary nutrient source was organic matter adsorbed onto the surfaces of the abioseston granules plus bacteria and detritus.E. gracilis feeds on algae, showing a preference for green algae and diatoms.C. vicinus is also a prodigious consumer of algae in Lake Balaton, utilizing the whole size spectrum of phytoplankton. Concerning the trophic relationships between phytoplankton and zooplankton in Lake Balaton, that between diatoms and bothE. gracilis andC. vicinus is the most conspicouos. Convincing evidence for an extensive utilization of blue-green algae was not found. Though there is no firm evidence yet, it is likely that theDaphnia are dependent on organic matter adsorbed on the abioseston.     

Mixed populations of marine microalgae in continuous culture: Factors affecting species dominance and biomass productivity

Marine microalgae were grown in multispecies continuous cultures. Under carbon dioxide limitation, blue-green algae dominated. Under nitrate and light limitation, species dominance depended on the initial conditions. When the inoculum consisted primarily of blue-green algae with smaller amounts of other species, blue-green algae and pennate diatoms dominated. When the inoculum consisted of equal amounts of all species, green flagellates and pennate diatoms dominated. Green flagellates and blue-green algae were incompatible and never shared dominance. When nutrient limitations were overcome, the productivity of seawater was increased 100-fold before light limitation occurred. The productivity could be further increased by reducing photorespiration in the culture. The dilution rates studied (0.1, 0.2, and 0.4 day(-1)) had no effect on species dominance, nor did the higher dilution rates select for smaller cells. The maximum productivity occurred at a dilution rate of 0.2 day(-1). Temperature had the greatest effect on species dominance, with green flagellates, pennate diatoms, and blue-green algae dominating at 20 degrees C and only blue-green algae dominating at 35 degrees C. The productivity at 35 degrees C was lower than that at 20 degrees C because of the lower solubility of carbon dioxide at higher temperatures. At 10% salinity, green flagellates and pennate diatoms dominated. The productivity at this salinity was 50% that obtained at the salinity of seawater (3.5%). At 25% salinity, only the green flagellate, Dunaliella salina, survived at a productivity of 1% that obtained at the salinity of seawater.     

Process-based simulation library SALMO-OO for lake ecosystems. Part 2: Multi-objective parameter optimization by evolutionary algorithms

SALMO-OO represents an object-oriented simulation library for lake ecosystems that allows to determine generic model structures for certain lake categories. It is based on complex ordinary differential equations that can be assembled by alternative process equations for algal growth and grazing as well as zooplankton growth and mortality. It requires 128 constant parameters that are causally related to the metabolic, chemical and transport processes in lakes either estimated from laboratory and field experiments or adopted from the literature.An evolutionary algorithm (EA) was integrated into SALMO-OO in order to facilitate multi-objective optimization for selected parameters and to substitute them by optimum temperature and phosphate functions. The parameters were related to photosynthesis, respiration and grazing of the three algal groups diatoms, green algae and blue-green algae. The EA determined specific temperature and phosphate functions for same parameters for 3 lake categories that were validated by ecological data of six lakes from Germany and South Africa.The results of this study have demonstrated that: (1) the hybridization of ordinary differential equations by EA provide a sophisticated approach to fine-tune crucial parameters of complex ecological models, and (2) the multi-objective parameter optimization of SALMO-OO by EA has significantly improved the accuracy of simulation results for three different lake categories.     

Dynamics of phytoplankton growth in the Waikato River,North Island,New Zealand

The Waikato River (latitude 38°S, longitude 176°E, North Island, New Zealand) is overwhelming y dominated by diatoms (mainly Melosira species) while blue-green and green algae are of minor importance. Both laboratory and in situ nutrient enrichment experiments showed enhanced growth of natural and index blue-green and green algae by addition of phosphate and nitrate. These algae were also shown to require higher temperature and light intensity than the diatoms. On the other hand, Waikato River with its higher silica content, moderate range of temperature and running water habitat was more favourable an environment for diatoms.     

Effect of larval fish and nutrient enrichment on plankton dynamics in experimental ponds

The hypotheses that larval fish density may potentially affect phytoplankton abundance through regulating zooplankton community structure, and that fish effect may also depend on nutrient levels were tested experimentally in ponds with three densities of larval walleye, Stizostedion vitreum (0, 25, and 50 fish m^–3), and two fertilizer types (inorganic vs organic fertilizer). A significant negative relationship between larval fish density and large zooplankton abundance was observed despite fertilizer types. Larval walleye significantly reduced the abundances of Daphnia, Bosmina, and Diaptomus but enhanced the abundance of various rotifer species (Brachionus, Polyarthra, and Keratella). When fish predation was excluded, Daphnia became dominant, but Daphnia grazing did not significantly suppress blue-green algae. Clearly, larval fish can be an important regulator for zooplankton community. Algal composition and abundance were affected more by fertilizer type than by fish density. Inorganic fertilizer with a high N:P ratio (20:1) enhanced blue-green algal blooms, while organic fertilizer with a lower N:P ratio (10:1) suppressed the abundance of blue-green algae. This result may be attributed to the high density of blue-green algae at the beginning of the experiment and the fertilizer type. Our data suggest that continuous release of nutrients from suspended organic fertilizer at a low rate may discourage the development of blue-green algae. Nutrient inputs at a low N:P ratio do not necessarily result in the dominance of blue-green algae.     

DETECTION OF LIMITING OR SURPLUS NITROGEN IN ALGAE AND AQUATIC WEEDS1

The rate of NH₄+-N absorption by algae and aquatic weeds in the dark has been shown to be 4-5 times greater for plants which are N-limited as compared to plants with sufficient available N. Eight species of green algae, 2 blue-green algae, 2 diatoms, and 3 aquatic weeds were used to demonstrate the usefulness of the test in determining if available N was in surplus or limited supply in a particular environment. The test was shown not to differentiate between blue-green algae capable of fixing N (4 species) from media with NO₃-N or without combined nitrogen. The factors influencing the results of NH4⁺-N absorption tests have been investigated. In order to differentiate between plants with sufficient available N and those which are N limited, the rate of NH₄⁺-N absorption (0.1 mg N) over 1-hr incubation in the dark by 10-20 mg of algae or aquatic weed tissues is measured. The relatively simple analysis for NH₄⁺-N in the samples makes it very easy to follow the changing N nutrition of plants in cultures with a limited N supply or in the presence of possible N sources.