Implication of community-level ecophysiological parameterization to modelling ecosystem productivity:a case study across nine contrasting forest sites in eastern China

Parameterization is a critical step in modelling ecosystem dynamics.However,assigning parameter values can be a technical challenge for structurally complex natural plant communities;uncertainties in model simulations often arise from inappropriate model parameterization.Here we compared five methods for defining community-level specific leaf area(SLA)and leaf C:N across nine contrasting forest sites along the North-South Transect of Eastern China,including biomass-weighted average for the entire plant community(AP_BW)and four simplified selective sampling(biomass-weighted average over five dominant tree species[5DT_BW],basal area weighted average over five dominant tree species[5DT_AW],biomass-weighted average over all tree species[AT_BW]and basal area weighted average over all tree species[AT_AW]).We found that the default values for SLA and leaf C:N embedded in the Biome-BGC v4.2 were higher than the five computational methods produced across the nine sites,with deviations ranging from 28.0 to 73.3%.In addition,there were only slight deviations(<10%)between the whole plant community sampling(AP_BW)predicted NPP and the four simplified selective sampling methods,and no significant difference between the predictions of AT_BW and AP_BW except the Shennongjia site.The findings in this study highlights the critical importance of computational strategies for community-level parameterization in ecosystem process modelling,and will support the choice of parameterization methods.

A New Result on Regular Designs under Baseline Parameterization

The study on designs for the baseline parameterization has aroused attention in recent years. This paper focuses on two-level regular designs for the baseline parameterization. A general result on the relationship between K-aberration and word length pattern is developed.

Parameterization Transfer for a Planar Computational Domain in Isogeometric Analysis

In this paper,we propose a parameterization transfer algorithm for planar domains bounded by B-spline curves,where the shapes of the planar domains are similar.The domain geometries are considered to be similar if their simplified skeletons have the same structures.One domain we call source domain,and it is parameterized using multi-patch B-spline surfaces.The resulting parameterization is C1 continuous in the regular region and G1 continuous around singular points regardless of whether the parameterization of the source domain is C1/G1 continuous or not.In this algorithm,boundary control points of the source domain are extracted from its parameterization as sequential points,and we establish a correspondence between sequential boundary control points of the source domain and the target boundary through discrete sampling and fitting.Transfer of the parametrization satisfies C1/G1 continuity under discrete harmonic mapping with continuous constraints.The new algorithm has a lower calculation cost than a decomposition-based parameterization with a high-quality parameterization result.We demonstrate that the result of the parameterization transfer in this paper can be applied in isogeometric analysis.Moreover,because of the consistency of the parameterization for the two models,this method can be applied in many other geometry processing algorithms,such as morphing and deformation.

Evaluating Parameterizations for Turbulent Fluxes over the Landfast Sea-Ice Surface in Prydz Bay, Antarctica

It is crucial to appropriately determine turbulent fluxes in numerical models.Using data collected in East Antarctica from 8 April to 26 November 2016,this study evaluates parameterization schemes for turbulent fluxes over the landfast seaice surface in five numerical models.The Community Noah Land Surface Model with Multi-Parameterizations Options(Noah_mp)best replicates the turbulent momentum flux,while the Beijing Climate System Model(BCC_CSM)produces the optimum sensible and latent heat fluxes.In particular,two critical issues of parameterization schemes,stability functions and roughness lengths,are investigated.Sensitivity tests indicate that roughness lengths play a decisive role in model performance.Based on the observed turbulent fluxes,roughness lengths over the landfast sea-ice surface are calculated.The results,which can provide a basis for setting up model parameters,reveal that the dynamic roughness length(z0m)increases with the increase of frictional velocity(u*)when u*≤0.4 m s^(−1) and fluctuates around 10^(−3 )m when u*>0.4 m s^(−1);thermal roughness length(z0t)is linearly related to the temperature gradient between air and sea-ice surface(ΔT)with a relation of lg(z0t)=−0.29ΔT−3.86;and the mean water vapor roughness length(z0q)in the specific humidity gradient(Δq)range ofΔq≤−0.6 g kg^(−1) is 10^(−6) m,3.5 times smaller than that in the range ofΔq˃−0.6 g kg^(−1).

Impact of seepage flow on sediment resuspension by internal solitary waves:parameterization and mechanism

Sediment incipient motion is the first step in sediment resuspension.Previous studies ignored the effect of seepage flow on the mobility of sediment particles and simplified the seabed surface as a rigid boundary.A flume experiment was designed to innovatively divide the seabed into two parts to control the dynamic response of the seabed and control the seepage conditions.In the experiment,the seabed sediments and the amplitude of internal solitary waves(ISWs)were changed to compare and analyze the impact of seepage flow on the sediment resuspension by shoaling ISWs.Moreover,parametric research and verification were carried out.Results indicate that seepage flow can greatly influence fine sand,promote sediment resuspension,and increase the amount of suspension by two times on average.However,seepage flow had a little effect on the suspension of clayey silt and sandy silt.Besides,seepage force was added to the traditional gravity,drag force,and uplift force,and the parameterization of threshold starting shear stress of coarse-grained sediments was developed.The results of this parameterization were verified,and seepage force was critical to parameterization.The threshold starting shear stress was reduced by 54.6%after increasing the seepage force.The physical mechanism of this process corresponded to the vertical reciprocating transient seepage in and out the seabed interface caused by the wave-induced transient excess pore water pressure.This quantitative study on seepage flow for shear stress of coarse-grained sediments induced by ISWs is critical to geohazard assessment.

The Importance of the Shape Parameter in a Bulk Parameterization Scheme to the Evolution of the Cloud Droplet Spectrum during Condensation

The shape parameter of the Gamma size distribution plays a key role in the evolution of the cloud droplet spectrum in the bulk parameterization schemes. However, due to the inaccurate specification of the shape parameter in the commonly used bulk double-moment schemes, the cloud droplet spectra cannot reasonably be described during the condensation process. Therefore, a newly-developed triple-parameter condensation scheme with the shape parameter diagnosed through the number concentration, cloud water content, and reflectivity factor of cloud droplets can be applied to improve the evolution of the cloud droplet spectrum. The simulation with the new parameterization scheme was compared to those with a high-resolution Lagrangian bin scheme, the double-moment schemes in a parcel model, and the observation in a 1.5D Eulerian model that consists of two cylinders. The new scheme with the shape parameter varying with time and space can accurately simulate the evolution of the cloud droplet spectrum. Furthermore, the volume-mean radius and cloud water content simulated with the new scheme match the Lagrangian analytical solutions well, and the errors are steady, within approximately 0.2%.

Performance Evaluation of Three Parameterizations on Internal Tidal Mixing in the Northern Pacific

The accurate assessment of the energy dissipation of internal tides(ITs)is of great importance because ITs contribute significantly to abyssal mixing.Thus,in this study,the IT-driven dissipation and diapycnal diffusion in the northern Pacific are esti-mated using parameterizations proposed by St.Laurent et al.(2002),Koch-Larrouy et al.(2007),and de Lavergne et al.(2020)(hereaf-ter referred to as LSJ02,KL07,and dL20,respectively).The performances of the three parameterizations are evaluated by comparing the calculated results with fine structure observations.In particular,the dissipation estimated by LSJ02 parameterization shows a bottom-intensified characteristic,with the patterns showing good agreement with the observations near seamounts.Moreover,43%of the results calculated using the LSJ02 parameterization have errors lower than one order of magnitude in the generation sites of ITs.Meanwhile,the strongest dissipation estimated by the KL07 parameterization shifts to the thermocline,with the results showing the highest level of consistency with observations in the generation sites.The proportion of results with errors lower than one order of magnitude is 80.7%.Furthermore,the results calculated by dL20 parameterization agree well with the observations in the upper and middle layers,with the parameterization showing an accurate estimation of the remote dissipation.The percentages of the errors lower than one order of magnitude between the dL20 parameterization and observations account for 77.1%and 88.7%in the genera-tion sites and far-field regions,respectively.

Parameterization Method of Wind Drift Factor Based on Deep Learning in the Oil Spill Model

Oil spill prediction is critical for reducing the detrimental impact of oil spills on marine ecosystems,and the wind strong-ly influences the performance of oil spill models.However,the wind drift factor is assumed to be constant or parameterized by linear regression and other methods in existing studies,which may limit the accuracy of the oil spill simulation.A parameterization method for wind drift factor(PMOWDF)based on deep learning,which can effectively extract the time-varying characteristics on a regional scale,is proposed in this paper.The method was adopted to forecast the oil spill in the East China Sea.The discrepancies between predicted positions and actual measurement locations of the drifters are obtained using seasonal statistical analysis.Results reveal that PMOWDF can improve the accuracy of oil spill simulation compared with the traditional method.Furthermore,the parameteriza-tion method is validated with satellite observations of the Sanchi oil spill in 2018.

Impact of Revised Trigger and Closure of the Double-Plume Convective Parameterization on Precipitation Simulations over East Asia

A double-plume convective parameterization scheme is revised to improve the precipitation simulation of a global model(Global-to-Regional Integrated Forecast System;GRIST).The improvement is achieved by considering the effects of large-scale dynamic processes on the trigger of deep convection.The closure,based on dynamic CAPE,is improved accordingly to allow other processes to consume CAPE under the more restricted convective trigger condition.The revised convective parameterization is evaluated with a variable-resolution model setup(110–35 km,refined over East Asia).The Atmospheric Model Intercomparison Project(AMIP)simulations demonstrate that the revised convective parameterization substantially delays the daytime precipitation peaks over most land areas,leading to an improved simulated diurnal cycle,evidenced by delayed and less frequent afternoon precipitation.Meanwhile,changes to the threshold of the trigger function yield a small impact on the diurnal amplitude of precipitation because of the consistent setting of dCAPE-based trigger and closure.The simulated mean precipitation remains reasonable,with some improvements evident along the southern slopes of the Tibetan Plateau.The revised scheme increases convective precipitation at the lower levels of the windward slope and reduces the large-scale precipitation over the upper slope,ultimately shifting the rainfall peak southward,which is in better agreement with the observations.

Skeleton-Based Volumetric Parameterizations for Lattice Structures

Lattice structures with excellent physical properties have attracted great research interest.In this paper,a novel volume parametric modeling method based on the skeleton model is proposed for the construction of threedimensional lattice structures.The skeleton model is divided into three types of nodes.And the corresponding algorithms are utilized to construct diverse types of volume parametric nodes.The unit-cell is assembled with distinct nodes according to the geometric features.The final lattice structure is created by the periodic arrangement of unit-cells.Several different types of volume parametric lattice structures are constructed to prove the stability and applicability of the proposed method.The quality is assessed in terms of the value of the Jacobian matrix.Moreover,the volume parametric lattice structures are tested with the isogeometric analysis to verify the feasibility of integration of modeling and simulation.

Parameterization and Explicit Modeling of Cloud Microphysics:Approaches, Challenges, and Future Directions

Cloud microphysical processes occur at the smallest end of scales among cloud-related processes and thus must be parameterized not only in large-scale global circulation models(GCMs)but also in various higher-resolution limited-area models such as cloud-resolving models(CRMs)and large-eddy simulation(LES)models.Instead of giving a comprehensive review of existing microphysical parameterizations that have been developed over the years,this study concentrates purposely on several topics that we believe are understudied but hold great potential for further advancing bulk microphysics parameterizations:multi-moment bulk microphysics parameterizations and the role of the spectral shape of hydrometeor size distributions;discrete vs“continuous”representation of hydrometeor types;turbulence-microphysics interactions including turbulent entrainment-mixing processes and stochastic condensation;theoretical foundations for the mathematical expressions used to describe hydrometeor size distributions and hydrometeor morphology;and approaches for developing bulk microphysics parameterizations.Also presented are the spectral bin scheme and particle-based scheme(especially,super-droplet method)for representing explicit microphysics.Their advantages and disadvantages are elucidated for constructing cloud models with detailed microphysics that are essential to developing processes understanding and bulk microphysics parameterizations.Particle-resolved direct numerical simulation(DNS)models are described as an emerging technique to investigate turbulence-microphysics interactions at the most fundamental level by tracking individual particles and resolving the smallest turbulent eddies in turbulent clouds.Outstanding challenges and future research directions are explored as well.

A 3D attention U-Net network and its application in geological model parameterization

To solve the problems of convolutional neural network–principal component analysis(CNN-PCA)in fine description and generalization of complex reservoir geological features,a 3D attention U-Net network was proposed not using a trained C3D video motion analysis model to extract the style of a 3D model,and applied to complement the details of geologic model lost in the dimension reduction of PCA method in this study.The 3D attention U-Net network was applied to a complex river channel sandstone reservoir to test its effects.The results show that compared with CNN-PCA method,the 3D attention U-Net network could better complement the details of geological model lost in the PCA dimension reduction,better reflect the fluid flow features in the original geologic model,and improve history matching results.

Effect of Different Microphysical Parameterizations on the Simulations of a South China Heavy Rainfall

A heavy rainfall event in south China was simulated by the Weather Research and Forecasting(WRF) model with three microphysics schemes, including the Morrison scheme, Thompson scheme, and Milbrandt and Yau scheme(MY), which aim to evaluate the capability to reproduce the precipitation and radar echo reflectivity features, and to evaluate evaluate their differences in microphysics and the associated thermodynamical and dynamical feedback. Results show that all simulations reproduce the main features crucial for rainfall formation. Compared with the observation, the MY scheme performed better than the other two schemes in terms of intensity and spatial distribution of rainfall. Due to abundant water vapor, the accretion of cloud droplets by raindrops was the dominant process in the growth of raindrops while the contribution of melting was a secondary effect. Riming processes, in which frozen hydrometeors collect cloud droplets mainly, contributed more to the growth of frozen hydrometeors than the Bergeron process. Extremely abundant snow and ice were produced in the Thompson and MY schemes respectively by a deposition process. The MY scheme has the highest condensation and evaporation, but the lowest deposition. As a result, in the MY scheme, the enhanced vertical gradient of condensation heating and evaporation cooling at low levels produces strong positive and weak negative potential vorticity in Guangdong, and may favor the formation of the enhanced rainfall center over there.

结构化网格快速生成工具的开发及其在冶金模拟仿真中的应用

结构化网格在模拟计算高精度、复杂问题时具有更高的计算准确度和更好的收敛性,但其划分难度大且耗时较长,如何快速生成高质量结构化网格是进行CFD(Computational fluid dynamics)等模拟计算领域的前沿问题.根据冶金容器的几何结构特征,引入了参数一体化建模+网格划分的理念,利用Python脚本快速批处理数据和开源CFD软件OpenFOAM的优势,成功开发了适用于冶金反应器的网格快速生成工具.以钢包–中间包–结晶器及其附属连接装置为例,进行了案例分析测试.结果表明,该工具软件生成的结构化网格在扭曲度和正交性等方面质量较高,可以满足各类模拟仿真的需求.该工具在解决冶金容器几何结构更新和网格分布调整方面相比于传统前处理方法可以在几十秒内快速生成高质量结构化网格.此外,开发了便于用户使用的图形用户界面,界面中仅需输入关键的几何结构和网格参数,即可对冶金容器快速便捷地生成相应的结构化网格,有效提升了前处理的简易度和效率.这一工具软件有望应用于冶金数字化和智能化转型发展的多种场景,并为其提供有效的软件支撑.

微细金属Z-pin对复合材料开孔板压缩性能的影响

通过开孔板压缩试验和建立的参数化多尺度有限元模型,获得微细(φ0.11 mm)金属Z-pin植入体积分数和排布方式对开孔板压缩力学性能和失效行为的影响规律.采用离散实体单元代表Z-pin,选用3D Hashin失效准则判断面内起始损伤,可以有效地模拟结构失效过程中扭结现象的不稳定扩展.结果表明,所有加Z-pin开孔板的压缩强度均低于无Z-pin试样.随着Z-pin植入体积分数的增加,Z-pin与层合板之间的桥联作用增强,加Z-pin开孔层合板压缩强度增加,开孔周围分层损伤区域受到抑制,损伤区域面积最高减小了67%.在相同的体积分数下,Z-pin排布变化对开孔板压缩强度没有显著影响.加Z-pin开孔板有限元模型的模拟结果与试验结果之间的最大相对误差为8.6%.

不同边界层参数化方案在中国西南地区东部高分辨率气候模拟中的敏感性研究

为了进一步评估和提高区域模式对西南地区东部高分辨率气候的模拟能力,利用WRF模式,采用多种边界层参数化方案(下称“不同方案”)对西南地区东部1998—2019年夏季降水和气温进行双重嵌套模拟(外层为D01,内层为D02)。对比不同方案模拟结果表明:多年平均降水量在D01中基本为湿偏差;D02中在四川盆地和重庆低海拔地区为干偏差,湿偏差主要位于贵州和重庆的城口、石柱和武隆一带的地形复杂区;总体上D01中ACM2方案误差最小,D02中MYJ方案误差最小。对多年平均气温的模拟在D01中除了四川盆地一带为暖偏差外其余大部地区基本为冷偏差,D02中大部地区为暖偏差;总体上D01和D02中MYJ方案误差最小,YSU方案最大。对于降水量和平均气温年际变化的模拟技巧在D01和D02中相对较高的地区均集中在重庆中西部和湖北大部地区;降水量总体为YSU方案最高,MYJ方案最低;平均气温总体为MYJ方案最高,ACM2方案最低。因此,提升模式分辨率至对流尺度后对不同气象要素模拟技巧最优的方案存在差异,需根据业务情况选择适合本地的参数化方案。

基于隐式参数化模型的地铁车体轻量化设计

为实现地铁新车型前期快速的概念设计和性能评估,引用“分析驱动设计”理念对重庆6号线地铁的MP车体进行了隐式参数化建模;并录入车体的17个主要钣金件进行设计分析。以板厚为设计变量,筛选出模态和刚度对质量相对灵敏度较大的7个车体钣金件,进行基于响应面法的多目标轻量化优化设计;车体的刚度、模态等多项性能指标达到了良好的轻量化效果。研究结果表明:所引入“分析驱动设计”理念的车体隐式参数化建模及基于响应面法的多目标优化方法能应用于车体结构的轻量化设计,具有较好的工程指导意义。

基于双线性RepVGG注意力网络的花卉分类

为进一步提高花卉分类的准确率,在对双线性卷积神经网络、RepVGG及注意力机制进行研究的基础上,提出一种基于双线性RepVGG注意力机制的网络模型。首先利用RepVGG网络替换原始的特征提取网络VGG,以提高对花卉主要特征的提取能力;然后在两个RepVGG网络中分别引入通道注意力及空间注意力机制,并利用两个RepVGG网络外积后生成的高维双线性特征,来提取花卉的细粒度特征;最后通过结构重参数化,将RepVGG的各层转换为单路结构,以提高模型推理的速度。实验结果表明,在增强的Oxford-102数据集上,新模型与原始模型及常见模型相比,其推理速度及分类准确率均有较大的提升,与未引入注意力前相比,分类准确率也有一定的提升。

轻量级重参数化的遥感图像超分辨率重建网络设计

针对当前基于深度学习的遥感图像超分辨率重建模型部署时对硬件要求较高,本文设计了一种轻量级基于重参数化的残差特征遥感图像超分辨率重建网络。首先,采用重参数化方法设计了一种残差局部特征模块,以有效地提取图像局部特征;同时考虑到图像内部出现的相似特征,设计了一个轻量级的全局上下文模块对图像的相似特征进行关联以提升网络的特征表达能力,并通过调整该模块的通道压缩倍数来减少模型的参数量和改善模型的性能;最后,在上采样模块前使用多层特征融合模块聚合所有的深度特征,以产生更全面的特征表示。在UC Merced遥感数据集上进行测试,该算法在遥感图像3倍超分辨率下的参数量为539 K,峰值信噪比为30.01 dB,结构相似性为0.8449,模型的推理时间为0.010 s;而HSENet算法的参数量为5470 K,峰值信噪比为30.00 dB,结构相似性为0.8420,模型的推理时间为0.059 s。实验结果表明,该算法相比HSENet算法,参数量更少,运行速度较快,且峰值信噪比与结构相似性也有一定的提高。在DIV2K自然图像数据集上进行测试,该算法的峰值信噪比和结构相似性相比其他算法也有一定的优势,表明该算法的泛化能力较强。

联合结构重参数和YOLOv5的航拍红外目标检测

无人机进行红外航拍目标检测在交通、农业和军事等方面有着广泛应用。该领域的主要挑战有目标较小、相互遮挡、非刚体形变大以及红外成像纹理信息少、边缘特征弱等。针对以上问题,基于YOLOv5和结构重参数化优化思想,提出了一种针对航拍场景的目标检测模型Rep-YOLO。首先,在主干网络中引入RepVGG模块,提升模型特征提取能力;在模型推理时对RepVGG模块的多分支进行结构重参数化,减少网络分支和结构复杂度。其次,结合数据特征,改进检测网络颈部的路径聚合网络,提升检测算法在机载平台的精度-速度均衡能力。最后,在两个公开红外数据集进行对比实验,表明该算法的有效性。以南航ComNet航拍数据集为例,统计结果显示主要检测指标各类平均精度(mean Average Precision,mAP)提升5.9%,同时参数量和模型大小分别减少约29.7%和23.2%。另外,对Rep-YOLO在典型机载平台Jetson Nano上进行了模型部署验证,为航拍场景的检测算法改进和实际应用提供了可靠的技术支撑。