wavewatch3.v4.18安装使用教程(浙江海洋大学修改于20160322).docx

Wavewatch3.v4.18在服务器上安装使用教程浙江海洋学院 盛叶新鉴于wavewatch3.v4.18的安装仅限于一本全英文的mannul手册，网络上没有任何有帮助的安装教程以及使用教程。因此，我于2015年10月在浙江海洋大学学习wavewatch3.v4.18时，在中国海大的两位师兄的帮助下成功安装了模式，仅针对我们学校的某个服务器安装的一些问题和步骤进行下列总结。一、环境变量设置安装ww3之前需要加载编译器并设置环境变量，不同的服务器有不同的加载方式，一般情况下使用module list 查看当前已加载编译器，module avail 查看可加载编译器，module load name 加载name编译器。安装ww3需要加载ifort、icc和openmpi-Intel（注意：使用openmpi-Intel是因为我们的集群上安装的是这个，其他mpi有关的编译器没有使用过，所以不太清楚）三个编译器。但我使用的服务器上不可使用module命令（具体原因和服务器配置有关，我没有权限深究，此处不做详解），因此我们设置环境变量文件.bashrc 文件。在修改之前，使用which ifort 、which icc和which mpirun查看三个编译的安装目录。根目录下输入vi .bashrc 添加如下代码：export PATH=$PATH:/(使用查找出来的ifort和icc安装目录)/bin/(可能发生ifort和icc是同一个目录，此处填写该目录)export PATH=$PATH:/(使用查找出来的mpi安装目录)/binexport LD_LIBRARY_PATH=/(使用查找出来的mpi安装目录)/lib/:$LD_LIBRARY_PATH此时环境变量则修改完毕，记得使用source .bashrc加载设置好的环境变量，然后继续完成安装。二、ww3安装和编译1.安装个人建议直接上传ww3的压缩文件，这样可以节约时间，也可先在windows下解压后在上传，最好是建立一个新的文件夹，这样不会发生文件找不到的情况。我提供的是4.18版本，文件夹示例：/public/home/shao002113/wwatch3.v4.18上传完毕后，按如下步骤进行：（注意空格）tar xvf wwatch3.v4.18 -C /public/home/shao002113/wwatch3.v4.18cd wwatch3.v4.18chmod +x install_ww3_tar./install_ww3_tar接下来进入安装了，他会问你一些问题，让你选择（y/n）的时候基本都是y。有一个是选择安装模式，一个是L一个是G，安装选择G，然后有些设置需要修改，如下：Printer for listingsprinter:printerCompiler for aux. f77: ifortCompiler for aux. cc: iccScratch space /tmp：这是一个关于tmp文件夹的设置，第一次安装不用修改Save source code files: yesSave listing files: yes之后出现的一律选择yes，会显示安装结束end。2.添加环境变量在根目录下输入命令：vi .bashrc在文本中添加：export PATH=$PATH:/(ww3完整安装目录)/binexport PATH=$PATH:/(ww3完整安装目录)/exe保存并退出。根目录下输入命令：source .bashrc（这是修改.bashrc后的必要步骤）。验证路径是否添加成功，根目录下输入 which w3_make.若成功修改，则会出现ww3完整目录。3.修改comp、link、switch进入bin文件夹，将comp.Intel和link.Intel复制成comp和link即：cp comp.Intel compcp link.Intel link使用vi编辑器修改comp和link。修改如下：comp中（大约96行处）修改成如下if $mpi_mod = yes then comp=mpiifort which mpiifort 1 /dev/null 2 /dev/null OK=$? if $OK != 0 then comp=mpif90 fi else comp=ifort filink中（大约104行）修改成如下if $mpi_mod = yes then comp=mpiifort which mpiifort 1 /dev/null 2 /dev/null OK=$? if $OK != 0 then comp=mpif90 fi opt=-O3 -o $prog else comp=ifort fiswitch中把SHRD改为MPI DIST(注意此处空格，否则会出错)。三个文件修改完毕后给予执行权限。chmod +u comp link switch4.运行make_MPI直接输入命令make_MPI即可。如果没有错误显示则编译成功，一般会出现两个错误，但不影响模式的运行计算，到此ww3安装编译成功。三、数据提取WW3对于输入数据文件的类型没有固定要求，本人常用的是风场为dat文件，地形为ASCII文件。那么输入数据的格式有着一定的要求。对于风场来说，数据排列方式如下：第一行是时间，如2015年6月1日0点0时0分，则写为：20150601 000000（注意年月日和时分秒之间存在空格）。接下来是两个矩阵，第一个是风场u，第二个是风场v。接下来是地形，说是地形，其实就是水深，大于等于0的设为无效值就好。当地行存为ASCII文件时，务必知道其绘出的地形图是什么形式的，在后面的修改脚本出需要用到。四、脚本修改1.ww3_grid.inp$ - $ WAVEWATCH III Grid preprocessor input file $ - $ Grid name (C*30, in quotes)$ TEST GRID (GULF OF NOWHERE) (此处是使用的网格名称)$ Frequency increment factor and first frequency (Hz) - $ number of frequencies (wavenumbers) and directions, relative offset$ of first direction in terms of the directional increment -0.5,0.5.$ In versions 1.18 and 2.22 of the model this value was by definiton 0,$ it is added to mitigate the GSE for a first order scheme. Note that$ this factor is IGNORED in the print plots in ww3_outp.$ 1.1 0.04118 25 24 0.（频率间隔 初始频率 波速 波向）$ Set model flags - $ - FLDRY Dry run (input/output only, no calculation).$ - FLCX, FLCY Activate X and Y component of propagation.$ - FLCTH, FLCK Activate direction and wavenumber shifts.$ - FLSOU Activate source terms.$ F T T T F T （上方六项的开关）$ Set time steps - $ - Time step information (this information is always read)$ maximum global time step, maximum CFL time step for x-y and$ k-theta, minimum source term time step (all in seconds).$ 900. 950. 900. 300. （第一个数和第三个数一般是第二个数的两倍，第二个数使用程序计算，最后一个一般为30）$ Start of namelist input section - $ Starting with WAVEWATCH III version 2.00, the tunable parameters$ for source terms, propagation schemes, and numerics are read using$ namelists. Any namelist found in the folowing sections up to the$ end-of-section identifier string (see below) is temporarily written$ to ww3_grid.scratch, and read from there if necessary. Namelists$ not needed for the given switch settings will be skipped$ automatically, and the order of the namelists is immaterial.$ As an example, namelist input to change SWELLF and ZWND in the$ Tolman and Chalikov input would be$ &SIN2 SWELLF = 0.1, ZWND = 15. / （此处均属于对源函数的调整，手册有相关介绍）$ Define constants in source terms - $ Stresses - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -$ TC 1996 with cap : Namelist FLX3$ CDMAX : Maximum allowed CD (cap)$ CTYPE : Cap type :$ 0: Discontinuous (default).$ 1: Hyperbolic tangent.$ Hwang 2011 : Namelist FLX4$ CDFAC : re-scaling of drag$ Linear input - - - - - - - - - - - - - - - - - - - - - - - - - - - -$ Cavaleri and M-R : Namelist SLN1$ CLIN : Proportionality constant.$ RFPM : Factor for fPM in filter.$ RFHF : Factor for fh in filter.$ Exponential input - - - - - - - - - - - - - - - - - - - - - - - - -$ WAM-3 : Namelist SIN1$ CINP : Proportionality constant.$ Tolman and Chalikov : Namelist SIN2$ ZWND : Height of wind (m).$ SWELLF : swell factor in (n.nn).$ STABSH, STABOF, CNEG, CPOS, FNEG :$ c0, ST0, c1, c2 and f1 in . (n.nn)$ through (2.65) for definition of$ effective wind speed (!/STAB2).$ WAM4 and variants : Namelist SIN3$ ZWND : Height of wind (m).$ ALPHA0 : minimum value of Charnock coefficient$ Z0MAX : maximum value of air-side roughness z0$ BETAMAX : maximum value of wind-wave coupling$ SINTHP : power of cosine in wind input$ ZALP : wave age shift to account for gustiness$ TAUWSHELTER : sheltering of short waves to reduce u_star$ SWELLFPAR : choice of swell attenuation formulation $ (1: TC 1996, 3: ACC 2008)$ SWELLF : swell attenuation factor$ Extra parameters for SWELLFPAR=3 only $ SWELLF2, SWELLF3 : swell attenuation factors $ SWELLF4 : Threshold Reynolds number for ACC2008$ SWELLF5 : Relative viscous decay below threshold$ Z0RAT : roughness for oscil. flow / mean flow $ BYDRZ input : Namelist SIN6$ SINA0 : factor for negative input$ SINU10 : wind speed scaling option$ Nonlinear interactions - - - - - - - - - - - - - - - - - - - - - - -$ Discrete I.A. : Namelist SNL1$ LAMBDA : Lambda in source term.$ NLPROP : C in sourc term. NOTE : default$ value depends on other source$ terms selected.$ KDCONV : Factor before kd in Eq. (n.nn).$ KDMIN, SNLCS1, SNLCS2, SNLCS3 :$ Minimum kd, and constants c1-3$ in depth scaling function.$ Exact interactions : Namelist SNL2$ IQTYPE : Type of depth treatment$ 1 : Deep water$ 2 : Deep water / WAM scaling$ 3 : Shallow water$ TAILNL : Parametric tail power.$ NDEPTH : Number of depths in for which$ integration space is established.$ Used for IQTYPE = 3 only$ Namelist ANL2$ DEPTHS : Array with depths for NDEPTH = 3$ Gen. Multiple DIA : Namelist SNL3$ NQDEF : Number of quadruplets.$ MSC : Scaling constant m.$ NSC : Scaling constant N.$ KDFD : Deep water relative filter depth,$ KDFS : Shallow water relative filter depth,$ Namelist ANL3$ QPARMS : 5 x NQDEF paramaters describing the$ quadruplets, repeating LAMBDA, MU, DT12.$ Cdeep and Cshal. See examples below.$ Traditional DIA setup (default):$ &SNL3 NQDEF = 1, MSC = 0.00, NSC = -3.50 /$ &ANL3 QPARMS = 0.250, 0.000, -1.0, 0.1000E+08, 0.0000E+00 /$ GMD3 from 2010 report (G13d in later paper) :$ &SNL3 NQDEF = 3, MSC = 0.00, NSC = -3.50 /$ &ANL3 QPARMS = 0.126, 0.000, -1.0, 0.4790E+08, 0.0000E+00 ,$ 0.237, 0.000, -1.0, 0.2200E+08, 0.0000E+00 ,$ 0.319, 0.000, -1.0, 0.1110E+08, 0.0000E+00 /$ G35d from 2010 report:$ &SNL3 NQDEF = 5, MSC = 0.00, NSC = -3.50 /$ &ANL3 QPARMS = 0.066, 0.018, 21.4, 0.170E+09, 0.000E+00 ,$ 0.127, 0.069, 19.6, 0.127E+09, 0.000E+00 ,$ 0.228, 0.065, 2.0, 0.443E+08, 0.000E+00 ,$ 0.295, 0.196, 40.5, 0.210E+08, 0.000E+00 ,$ 0.369, 0.226, 11.5, 0.118E+08, 0.000E+00 /$ Nonlinear filter based on DIA - - - - - - - - - - - - - - - - - - -$ Namelist SNLS$ A34 : Relative offset in quadruplet$ FHFC : Proportionality constants.$ DMN : Maximum relative change.$ FC1-3 : Constants in frequency filter.$ Dissipation - - - - - - - - - - - - - - - - - - - - - - - - - - - -$ WAM-3 : Namelist SDS1$ CDIS, APM : As in source term.$ Tolman and Chalikov : Namelist SDS2$ SDSA0, SDSA1, SDSA2, SDSB0, SDSB1, PHIMIN :$ Constants a0, a1, a2, b0, b1 and$ PHImin.$ WAM4 and variants : Namelist SDS3$ SDSC1 : WAM4 Cds coeffient$ MNMEANP, WNMEANPTAIL : power of wavenumber$ for mean definitions in Sds and tail $ SDSDELTA1, SDSDELTA2 : relative weights $ of k and k2 parts of WAM4 dissipation$ SDSLF, SDSHF : coefficient for activation of $ WAM4 dissipation for unsaturated (SDSLF) and $ saturated (SDSHF) parts of the spectrum$ SDSC2 : Saturation dissipation coefficient$ SDSC4 : Value of B0=B/Br for wich Sds is zero$ SDSBR : Threshold Br for saturation$ SDSP : power of (B/Br-B0) in Sds$ SDSBR2 : Threshold Br2 for the separation of $ WAM4 dissipation in saturated and non-saturated$ SDSC5 : coefficient for turbulence dissipation$ SDSC6 : Weight for the istropic part of Sds_SAT$ SDSDTH: Angular half-width for integration of B$ BYDRZ : Namelist SDS6$ SDSET : Select threshold normalization spectra$ SDSA1, SDSA2, SDSP1, SDSP2 :$ Coefficients for dissipation terms T1 and T2$ : Namelist SWL6$ SWLB1 : Coefficient for swell dissipation$ Bottom friction - - - - - - - - - - - - - - - - - - - - - - - - - -$ JONSWAP : Namelist SBT1$ GAMMA : As it says.$ Surf breaking - - - - - - - - - - - - - - - - - - - - - - - - - - -$ Battjes and Janssen : Namelist SDB1$ BJALFA : Dissipation constant (default = 1)$ BJGAM : Breaking threshold (default = 0.73)$ BJFLAG : TRUE - Use Hmax/d ratio only (default)$ FALSE - Use Hmax/d in Miche formulation$ Triad nonlinear interactions - - - - - - - - - - - - - - - - - - - -$ Lumped Triad Interaction (LTA) : Namelist STR1 (To be implemented)$ PTRIAD1 : Proportionality coefficient (default 0.05)$ PTRIAD2 : Multiple of Tm01 up to which interaction $ is computed (2.5)$ PTRIAD3 : Ursell upper limit for computing$ interactions (not used, default 10.)$ PTRIAD4 : Shape parameter for biphase$ computation (0.2)$ PTRIAD5 : Ursell number treshold for computing $ interactions (0.01)$ Shoreline reflections - - - - - - - - - - - - - - - - - - - - - - - - $ ref. parameters : Namelist REF1 $ REFCOAST : Reflection coefficient at shoreline$ REFFREQ : Activation of freq-dependent ref. $ REFMAP : Scale factor for bottom slope map$ REFRMAX : maximum ref. coeffient (default 0.8) $ REFFREQPOW: power of frequency $ REFICEBERG: Reflection coefficient for icebergs$ REFSUBGRID: Reflection coefficient for islands$ REFCOSP_STRAIGHT: power of cosine used for $ straight shoreline$ Bound 2nd order spectrum and free IG - - - - - - - - - - - - - - - - - $ IG1 parameters : Namelist IG1$ IGMETHOD : 1: Hasselmann, 2: Krasitskii-Janssen$ IGADDOUTP : activation of bound wave correction $ in ww3_outp / ww3_ounp$ IGSOURCE : 1: uses bound waves, 2: empirical$ IGSTERMS : 0 : no source term in IG band$ IGMAXFREQ : maximum frequency of IG band$ IGEMPIRICAL: constant in empirical free IG source$ IGSWELLMAX: activates free IG sources for all freq.$ Propagation schemes - $ SMC grid propagation : Namelist PSMC$ CFLTM : Maximum CFL number for propagation.$ DTIME : Swell age (s) for diffusion term.$ LATMIN : Maximum latitude (deg) for GCT. $ RFMAXD : Maximum refraction turning (deg).$ &PSMC DTIME = 39600.0, LATMIN=85.0, RFMAXD = 36.0 /$ Propagation schemes - $ First order : Namelist PRO1$ CFLTM : Maximum CFL number for refraction.$ UQ/UNO with diffusion : Namelist PRO2$ CFLTM : Maximum CFL number for refraction.$ DTIME : Swell age (s) in garden sprinkler$ correction. If 0., all diffusion$ switched off. If small non-zero$ (DEFAULT !) only wave growth$ diffusion.$ LATMIN : Maximum latitude used in calc. of$ strength of diffusion for prop.$ UQ/UNO with averaging : Namelist PRO3$ CFLTM : Maximum CFL number for refraction.$ WDTHCG : Tuning factor propag. direction.$ WDTHTH : Tuning factor normal direction.$ Note that UQ and UNO schemes have no tunable parameters.$ All tuneable parameters are associated with the refraction$ limitation and the GSE alleviation.$ Unstructured grids - $ UNST paramete