Learning objectives / 学习目标
学完本课后,你应能:
- 从 vertical momentum balance 推导 hydrostatic relation 与 isothermal scale height,并解释 potential temperature 和 static stability;
- 从 rotating-frame equation 定义 Coriolis parameter、Rossby number,建立 geostrophic balance 并判断适用范围;
- 用 buoyancy restoring force 区分 stable oscillation、dry convection、internal gravity wave 和 baroclinic instability;
- 以 wind stress、Ekman transport、density contrast、mixed-layer heat budget 解释 ocean circulation 与 air–sea coupling;
- 说明 numerical weather prediction 中 observation、data assimilation、model integration 与 ensemble calibration 的角色;
- 用 Lyapunov/error-growth model 区分 chaos、model error、measurement uncertainty 与 probabilistic forecast skill。
Prerequisites / 先修知识
- pressure、density、ideal-gas law、adiabatic process 与 buoyancy;
- Solid-Earth Physics 中的 reference model、wave、inverse problem 与 model boundary;
- angular velocity、non-inertial frames 与 vector cross product;
- continuity、momentum/energy conservation、partial derivatives;
- probability distribution、mean/spread 与 numerical approximation。
Scope and conventions / 范围与约定
- local Cartesian frame:\(x\) 向 east、\(y\) 向 north、\(z\) 向 upward,\(\hat{\mathbf k}\) 向上。
- Earth rotation vector 为 \(\boldsymbol\Omega\);Coriolis acceleration 在 rotating frame 写为 \(-2\boldsymbol\Omega\times\mathbf u\)。在传统 horizontal approximation 中 \(f=2\Omega\sin\phi\)。
- pressure-gradient force per unit mass 为 \(-\nabla p/\rho\);geostrophic balance 只针对 large-scale, slowly varying, nearly horizontal interior flow。
- ocean heat flux \(Q_{net}>0\) 表示进入 ocean mixed layer;wind stress \(\boldsymbol\tau\) 是 atmosphere 对 ocean 的切向力 per area。
- dry, Boussinesq, hydrostatic 与 shallow-water models 都是层级中的简化;正文每次使用会声明,不能把一个模型的结论无条件迁移到 hurricane、boundary layer 或 equator。
Notation and modeling contract / 符号与建模契约
| Symbol | Meaning | SI / convention |
|---|---|---|
| \(p,\rho,T\) | pressure, density, temperature | Pa; \(\mathrm{kg\,m^{-3}}\); K |
| \(R_d,c_p\) | dry-air gas constant and specific heat | \(\mathrm{J\,kg^{-1}\,K^{-1}}\) |
| \(H\) | isothermal pressure scale height | m |
| \(\theta\) | dry potential temperature | K |
| \(N\) | Brunt–Väisälä frequency | \(\mathrm{s^{-1}}\) |
| \(f\) | Coriolis parameter \(2\Omega\sin\phi\) | \(\mathrm{s^{-1}}\) |
| \(Ro\) | Rossby number \(U/(\lvert f\rvert L)\) | 1 |
| \(\mathbf u_g\) | geostrophic velocity | \(\mathrm{m\,s^{-1}}\) |
| \(\boldsymbol\tau\) | surface wind stress | Pa |
| \(\mathbf M_E\) | Ekman volume transport per unit width | \(\mathrm{m^2\,s^{-1}}\) |
| \(Q_{net}\) | net heat flux into mixed layer | \(\mathrm{W\,m^{-2}}\) |
| \(\lambda_L\) | leading finite-time error-growth rate | \(\mathrm{s^{-1}}\) |
1. Hydrostatic atmosphere and stratification / 大气静力平衡与分层
Knowledge ID: P14-U03-L02-K01
vertical acceleration 在 synoptic-scale atmosphere 中通常远小于 \(g\),vertical momentum equation 的 leading balance 是
\[\frac{\partial p}{\partial z}=-\rho g.\]
这不是说 air parcels 永不 vertical accelerate;它说 pressure field 对 parcel weight 的大部分已近乎平衡,剩余小差驱动 vertical motion。结合 dry ideal-gas law \(p=\rho R_dT\),若 \(T\) 与 \(g\) 近似 constant:
\[p(z)=p_0e^{-z/H},\qquad H=\frac{R_dT}{g}.\]
图 1 以 thin slab 说明 upper/lower pressure-force difference 如何平衡 weight;箭长为 schematic。
图 2 直接以 normalized coordinates 绘制 \(p/p_0=e^{-z/H}\):经过一个 scale height,即 \(z/H=1\),pressure 降至 \(e^{-1}\)。真实 atmosphere 的 \(T(z)\) 变化使 local scale height 变化。
判断 parcel 被小幅抬升后会返回还是继续远离,需要把 parcel 的 adiabatic temperature change 与 environment stratification 比较。dry potential temperature
\[\theta=T\left(\frac{p_0}{p}\right)^{R_d/c_p}\]
是 dry adiabatic displacement 中近似守恒的变量。Boussinesq/dry approximation 下
\[N^2=\frac{g}{\theta}\frac{d\theta}{dz}.\]
\(N^2>0\) 表示 stable stratification,小位移产生 restoring buoyancy;\(N^2<0\) 表示 dry convective instability。moist atmosphere 需改用 moisture-dependent buoyancy/entropy variables,不能仅用 dry \(\theta\) 判定 saturated parcel。
图 3 对比 \(d\theta/dz\) 的 sign;parcel arrow 表示 displacement,黑箭头表示 subsequent buoyancy acceleration。
2. Rotation, Coriolis force, and geostrophic balance / 旋转、科里奥利力与地转平衡
Knowledge ID: P14-U03-L02-K02
在 Earth-fixed rotating frame,momentum equation 含 Coriolis acceleration:
\[\frac{D\mathbf u}{Dt} =-\frac1\rho\nabla p+\mathbf g -2\boldsymbol\Omega\times\mathbf u+\mathbf F_{fric}.\]
Coriolis force 不做功,因为 \(\mathbf u\cdot(-2\boldsymbol\Omega\times\mathbf u)=0\);它改变 velocity direction 而不直接改变 kinetic energy。traditional horizontal approximation 下,水平 acceleration 为 \(-f\hat{\mathbf k}\times\mathbf u\),\(f=2\Omega\sin\phi\);Northern Hemisphere moving parcel 向 motion 的右侧偏,Southern Hemisphere 向左。
图 4 在相同初始 northward motion 下示意 opposite deflection;这不是惯性系中出现的新相互作用。
当 horizontal pressure-gradient acceleration 与 Coriolis acceleration dominant 且 nearly cancel:
\[f\hat{\mathbf k}\times\mathbf u_g=-\frac1\rho\nabla_h p.\]
geostrophic wind/current 沿 isobars/isobaths-associated pressure contours,而不是直接从 high pressure 冲向 low pressure。Northern Hemisphere 中,低压在 flow 左侧;near surface friction 使 wind 穿越 isobars 指向较低 pressure。适用性可由
\[Ro=\frac{U}{\lvert f\rvert L}\]
衡量:\(Ro\ll1\) 支持 rotation-dominated leading balance。equator 附近 \(f\to0\)、small tornadoes、rapid acceleration 或 boundary layers 不能用 simple geostrophy。
图 5 以等压线 orientation 表示 geostrophic balance。标签 northward 指图上 flow direction,不代表页面坐标的 north axis;图为 schematic。
3. Convection, waves, and instabilities / 对流、波与不稳定性
Knowledge ID: P14-U03-L02-K03
stable stratification 中 displaced parcel 可作 buoyancy oscillation:
\[\frac{d^2\zeta}{dt^2}+N^2\zeta=0,\]
其中 \(\zeta\) 是 vertical displacement。若 disturbance 同时具有 horizontal/vertical structure,restoring buoyancy 与 inertia 形成 internal gravity waves。在 nonrotating, inviscid, uniformly stratified Boussinesq fluid 中,plane-wave dispersion 为
\[\omega^2=N^2\frac{k_h^2}{k_h^2+m^2},\]
\(k_h\) 与 \(m\) 分别为 horizontal、vertical wavenumber。此式显示 \(0<\omega<N\);rotation 加入 inertial lower bound,compressibility 和 boundaries 也会改变谱。
图 6 中 blue curves 是 material-density surfaces 的 perturbation,不是 individual parcel trajectories。
若 \(N^2<0\),small displacement 被放大,形成 dry convection;with moisture,latent heating 会改变 parcel buoyancy。另一重要机制是 baroclinic instability:horizontal temperature/density gradients 对应 sloping pressure/density surfaces 与 vertical shear,large-scale disturbances 可把 available potential energy 转成 eddy kinetic energy,形成 midlatitude weather systems。它不是“任何温差都会爆发”,而需要 rotation、stratification、shear、wavelength 与 boundary conditions 的适当组合。
shallow-water external gravity wave 的 long-wave speed
\[c=\sqrt{gH}\]
适用于 wavelength 远大于 fluid depth 且 vertical acceleration 可忽略的 layer;tsunami 的 ocean-basin propagation 可由此近似,但 coastal shoaling、dispersion 与 breaking 需要扩展模型。
图 7 是 baroclinic reservoir 的截面示意;真正 instability analysis 需 perturbation equations 与 boundary conditions。
4. Ocean circulation and air–sea coupling / 海洋环流与海气耦合
Knowledge ID: P14-U03-L02-K04
ocean circulation 由 wind stress、buoyancy fluxes(heat/freshwater)、tides 与 pressure gradients 驱动,又受 rotation、stratification、basin geometry、bottom topography 和 mixing 控制。near-surface turbulent stress 在 rotating ocean 中形成 Ekman spiral/transport;在 constant-\(f\)、steady, deep-enough boundary-layer idealization 下,vertically integrated volume transport per width 为
\[\mathbf M_E=\frac{\boldsymbol\tau\times\hat{\mathbf k}}{\rho_0 f}.\]
Northern Hemisphere transport 在 wind stress 右侧;Southern Hemisphere 因 \(f<0\) 反向。equator、coasts、shallow water 与 rapidly varying winds 需要完整 momentum and continuity equations。Ekman convergence/downwelling 与 divergence/upwelling 把 surface forcing 连接到 interior circulation。
图 8 左 panel 是 horizontal plan view:在 Northern Hemisphere,eastward wind stress 的 vertically integrated \(\mathbf M_E\) 指向 south,即 wind stress 右侧。右 panel 只用 vertical placement 索引 depth,三个圆分别代表不同深度的 horizontal plan view;这不是 downward transport,\(\mathbf M_E\) 也没有画在 depth 方向。
large-scale ocean 包含 wind-driven gyres、western boundary currents、equatorial currents 与 density-driven overturning。短语 thermohaline circulation 容易误导为 temperature/salinity 单独构成一条固定 conveyor;真实 overturning 是 wind、buoyancy transformation、mixing and eddies 共同维持的 three-dimensional circulation。
air–sea interface 交换 momentum、sensible heat、latent heat、radiation、freshwater 与 gases。若 mixed layer depth \(h\) 固定、horizontal transport/mixing 可忽略:
\[\rho_0c_ph\frac{dT_m}{dt}=Q_{net}.\]
真实 \(h(t)\)、entrainment、advection 与 penetrative shortwave radiation 会改变 response。海洋的巨大 heat capacity 使它既延缓 surface warming,也储存并重新释放 anomaly;这连接 planetary energy balance 与 regional weather。
图 9 是 coupled budgets 图:surface fluxes 改变 ocean state,ocean SST/currents 又反馈 boundary layer 与 atmosphere。
5. Weather prediction and chaotic sensitivity / 天气预报与混沌敏感性
Knowledge ID: P14-U03-L02-K05
numerical weather prediction(NWP)不是从一张天气图“外推云朵”。它先把 satellites、radar、radiosondes、aircraft、ships、buoys 与 surface stations 的不均匀 observations,通过 data assimilation 与 previous forecast/background 合成 dynamically consistent initial state;随后离散求解 momentum、mass、thermodynamic 与 constituent equations,并 parameterize unresolved cloud microphysics、radiation、turbulence 与 surface exchange。
chaotic sensitivity 指 initial states 的 small difference 可沿 unstable directions 增长。在有限时间、线性阶段可写 toy model
\[\lVert\delta\mathbf x(t)\rVert \approx\lVert\delta\mathbf x(0)\rVert e^{\lambda_Lt}.\]
这不等于“预报完全随机”:large-scale regimes、conserved balances 与 probabilities 仍有 skill;也不等于所有误差均来自 initial state,model bias、subgrid parameterization、observation error 与 finite resolution 同样重要。
ensemble prediction 用 perturbed initial conditions、model physics or multiple models 采样 uncertainty。ensemble spread 不是自动等于 calibrated probability:需 reliability、resolution、rank histograms 与 out-of-sample verification。单一 deterministic run 的精细小数不应被误读为 certainty。
图 10 显示 initial close trajectories 的 spread growth;blue envelope 不是保证 truth 必在其中,校准必须靠 verification。
Worked examples / 完整例题
Worked Example 1 — isothermal scale height 与 pressure
已知: \(R_d=287\ \mathrm{J\,kg^{-1}\,K^{-1}}\)、\(T=288\ \mathrm K\)、\(g=9.81\ \mathrm{m\,s^{-2}}\)、\(p_0=101.3\ \mathrm{kPa}\),求 \(z=5.00\ \mathrm{km}\) 的 pressure。
研究对象与边界: dry isothermal atmospheric column;constant \(g\),忽略 water vapor 与 weather systems。
参考系/坐标: \(z=0\) at reference level,\(+z\) upward。
模型理由: hydrostatic ideal-gas model 适合 scale estimate。
\[H=\frac{R_dT}{g} =\frac{(287)(288)}{9.81} =8.43\times10^3\ \mathrm m,\]
\[p(5.00\ \mathrm{km}) =(101.3\ \mathrm{kPa})e^{-5000/8426} =55.9\ \mathrm{kPa}.\]
单位/有效数字: \(R_dT/g\) gives metres;\(p\) 取三位有效数字。
符号/合理性: exponent dimensionless and negative;\(p/p_0=0.552\),随高度下降但不变负。
极限/解释: \(z\to0\) gives \(p_0\);真实 lapse rate 使 5 km value 与此 isothermal estimate 有偏差。
Worked Example 2 — 由 pressure gradient 求 geostrophic wind
已知: Northern Hemisphere midlatitude,\(f=1.00\times10^{-4}\ \mathrm{s^{-1}}\)、\(\rho=1.20\ \mathrm{kg\,m^{-3}}\),pressure 向 east 每 \(100\ \mathrm{km}\) 增加 \(100\ \mathrm{Pa}\)。
研究对象与边界: frictionless horizontal interior air parcel;steady geostrophic approximation。
参考系/坐标: \(+x\) east、\(+y\) north、\(+z\) up。
模型理由: horizontal scale large and \(Ro\ll1\) is assumed; pressure varies only in \(x\)。
\[\frac{\partial p}{\partial x} =\frac{100}{1.00\times10^5} =1.00\times10^{-3}\ \mathrm{Pa\,m^{-1}},\]
从 \(v_g=(\rho f)^{-1}\partial p/\partial x\) 得
\[v_g=\frac{1.00\times10^{-3}}{(1.20)(1.00\times10^{-4})} =8.33\ \mathrm{m\,s^{-1}},\qquad u_g=0.\]
单位/有效数字: pressure-gradient divided by \(\rho f\) gives \(\mathrm{m\,s^{-1}}\);三位有效数字。
方向检查: \(v_g>0\) 即 northward;向 east 的 high pressure 位于 northward flow 的右侧,符合 NH rule。
模型边界/解释: near surface friction 会使 flow 斜穿 isobars;equator \(f\to0\) 时此 formula 发散,表明 balance 失效而非 wind 无限大。
Worked Example 3 — static stability 与 buoyancy period
已知: \(\theta=300\ \mathrm K\),\(d\theta/dz=3.00\ \mathrm{K\,km^{-1}}\),\(g=9.81\ \mathrm{m\,s^{-2}}\)。
研究对象与边界: dry Boussinesq, uniformly stratified parcel,small vertical displacement。
参考系/坐标: \(z\) upward;\(\zeta>0\) upward。
模型理由: linear displacement equation 用于诊断 stable oscillation。
\[N^2=\frac{9.81}{300}\left(\frac{3.00}{1000}\right) =9.81\times10^{-5}\ \mathrm{s^{-2}},\]
\[N=9.90\times10^{-3}\ \mathrm{s^{-1}}, \qquad \mathcal T_N=\frac{2\pi}{N}=6.35\times10^2\ \mathrm s=10.6\ \mathrm{min}.\]
单位/有效数字: \(N^2\) is \(\mathrm{s^{-2}}\);period 三位有效数字。
sign/limit check: \(d\theta/dz>0\) gives \(N^2>0\) and oscillatory solution;gradient approaching zero makes restoring period diverge。
物理解释: \(10.6\) min 是 ideal parcel buoyancy period;moisture、shear、mixing 与 finite wavelength 会改变 observed waves。
Worked Example 4 — mixed-layer warming 与 forecast-error growth
已知: ocean mixed layer \(\rho_0=1025\ \mathrm{kg\,m^{-3}}\)、\(c_p=3990\ \mathrm{J\,kg^{-1}\,K^{-1}}\)、\(h=50.0\ \mathrm m\),持续 \(30.0\) days 的 \(Q_{net}=50.0\ \mathrm{W\,m^{-2}}\);同时 forecast temperature-state error 初值 \(\delta T_0=0.100\ \mathrm K\)、e-folding time \(\lambda_L^{-1}=2.00\) days,求两种 toy-model response。
研究对象与边界: (a) fixed-depth, horizontally isolated mixed layer;(b) linear finite-time error norm;两部分不是同一 conservation equation。
参考系/符号: \(Q_{net}>0\) into ocean;error 报告 magnitude。
模型理由: 分别量化 thermal inertia 与 chaotic amplification。
\[\Delta T_m =\frac{Q_{net}\Delta t}{\rho_0c_ph} =\frac{(50.0)(30.0\times86400)}{(1025)(3990)(50.0)} =0.634\ \mathrm K,\]
\[\delta T(6.00\ \mathrm{days}) =(0.100)e^{6.00/2.00} =2.01\ \mathrm K.\]
单位/有效数字: energy per area divided by areal heat capacity gives K;两结果三位有效数字。
sign/reasonableness: positive heat flux warms;three e-folds multiply error by \(e^3\approx20.1\)。
模型边界/解释: entrainment/advection can reduce or enhance \(0.634\) K;error exponential cannot continue beyond nonlinear saturation,且 ensemble calibration 不能由一个 \(\lambda_L\) 决定。
Lesson synthesis / 全课综合
大气—海洋问题可按以下 scale-aware balance workflow 处理:
- 声明 fluid parcel/system、Earth-fixed frame、horizontal/vertical coordinates 与 averaging scale;
- 写完整 conservation laws,再用 aspect ratio、\(Ro\)、\(N^2\)、Reynolds/Péclet 等量级判断 leading balances;
- vertical 先检查 hydrostatic,horizontal 再检查 geostrophic、inertial、frictional or advective balance;
- 判断 background stratification 是否 stable,若偏离平衡,识别 wave、convection 或 instability 的 restoring/energy source;
- 对 air–sea interface 建 momentum、heat、freshwater ledger,避免把 atmosphere/ocean 各自当 closed system;
- 预报时把 observation error、initial-condition uncertainty、model error 与 unresolved scales 分开,并用 ensembles 加 verification 输出 probability。
同一现象可跨多个 regime:hurricane core 的 \(Ro\) 不小、surface layer friction 重要,而远离中心的 synoptic environment 可近 geostrophic;ocean interior slow balance 与 thin boundary currents/mesoscale eddies 也不能用一个方程无差别描述。
Formula-and-decision table / 公式与决策表
| Question | Relation | Use when | Main limitation |
|---|---|---|---|
| Vertical pressure? | \(\partial p/\partial z=-\rho g\) | vertical acceleration small versus \(g\) | deep convection and acoustic motion need full equation |
| Isothermal pressure? | \(p=p_0e^{-z/H}\), \(H=R_dT/g\) | dry, constant \(T,g\) | real lapse rate and moisture |
| Dry stability? | \(N^2=(g/\theta)d\theta/dz\) | small dry Boussinesq displacement | saturated/moist processes change criterion |
| Rotation importance? | \(Ro=U/(\lvert f\rvert L)\) | compare inertia with Coriolis | \(f\to0\) near equator |
| Geostrophic flow? | \(f\hat{\mathbf k}\times\mathbf u_g=-\nabla_hp/\rho\) | steady, large-scale, weak-friction interior | boundary layers and rapid acceleration |
| Internal gravity wave? | \(\omega^2=N^2k_h^2/(k_h^2+m^2)\) | uniform, nonrotating Boussinesq fluid | rotation, shear and boundaries alter dispersion |
| Shallow-water wave? | \(c=\sqrt{gH}\) | wavelength much greater than depth | dispersion, shoaling and breaking |
| Ekman transport? | \(\mathbf M_E=(\boldsymbol\tau\times\hat{\mathbf k})/(\rho_0f)\) | steady constant-\(f\) rotating boundary layer | equator/coast/time dependence |
| Mixed-layer temperature? | \(\rho_0c_ph\,dT_m/dt=Q_{net}\) | fixed \(h\), negligible advection/entrainment | coupling and mixed-layer-depth changes |
| Early error growth? | \(\lVert\delta\mathbf x(t)\rVert\approx\lVert\delta\mathbf x_0\rVert e^{\lambda_Lt}\) | finite-time linear unstable regime | nonlinear saturation and state dependence |
Model boundaries / 模型边界
- Hydrostatic: excellent for many large-scale motions, but excludes vertically propagating sound and can fail in intense small-scale updrafts.
- Dry thermodynamics: potential-temperature test needs moisture-aware replacement for clouds and saturated convection.
- Traditional \(f\)-plane: assumes local latitude and neglects horizontal component of \(\boldsymbol\Omega\); equatorial/\(\beta\) effects require extension.
- Geostrophy: describes leading interior balance, not actual acceleration mechanism or surface cross-isobar flow.
- Boussinesq: density variation retained only in buoyancy; unsuitable across the full compressible atmosphere without anelastic/compressible models.
- Shallow water: depth-averaged; cannot resolve vertical shear or internal stratified modes unless layered.
- Ekman model: idealizes turbulent closure and constant coefficients; observed spirals/transport vary with stratification and waves.
- Mixed-layer slab: no entrainment/advection/variable depth; use only for first heat-capacity estimate.
- NWP ensemble: finite members sample a designed uncertainty distribution, not every possible model inadequacy.
Misconception clinic / 误区诊所
- “hydrostatic means air is motionless.” 错。horizontal wind 与 slow vertical motion 可同时存在;只需 vertical acceleration 对 \(g\) 小。
- “Coriolis force 让 moving parcel 加速并增加 kinetic energy。” 错。它与 velocity 垂直,不直接做功;pressure、buoyancy/friction 才改变 energy。
- “wind 总是从高压直吹低压。” 只在 friction/acceleration 重要时有明显 cross-isobar component;free atmosphere 常近 geostrophic along-isobar flow。
- “\(N^2>0\) 表示没有 vertical motion。” 错。它表示 small displacement 被 restoring,可支持 internal waves。
- “所有 cyclone 都是 Coriolis 造成的,洗手盆也一样。” 错。rotation relevance 由 \(Ro\) 与 initial/boundary asymmetry 决定;small drains 常 \(Ro\gg1\)。
- “thermohaline conveyor 是一条固定水带。” 错。overturning 是 3-D、time-dependent、wind–buoyancy–mixing coupled circulation。
- “ensemble mean 一定会发生,spread 就是精确概率。” 错。mean 可能不是任何 member trajectory;probability 需 calibrated verification。
- “chaos 意味 physics model 无用。” 错。chaos 限制 trajectory predictability,却不消除 balance、short-lead skill 与 probabilistic predictability。
Practice and retrieval / 练习与检索
- [K01] isothermal atmosphere 中 \(T\) 增加 \(10\%\)、\(g\) 不变,scale height 如何变化?固定 \(z>0\) 的 \(p/p_0\) 如何变化?
Answer 1
\(H=R_dT/g\) 增加 \(10\%\);\(p/p_0=e^{-z/H}\) 的 exponent magnitude 变小,所以 ratio 增大,pressure 随高度下降得更慢。
- [K01] \(\theta=290\ \mathrm K\)、\(d\theta/dz=-1.0\ \mathrm{K\,km^{-1}}\),求 \(N^2\) 的 sign 并判断 dry stability。
Answer 2
\(N^2=(9.81/290)(-0.001)=-3.38\times10^{-5}\ \mathrm{s^{-2}}<0\),在 dry Boussinesq criterion 下 convectively unstable;moisture/shear 尚未计入。
- [K02] \(U=10\ \mathrm{m\,s^{-1}}\)、\(L=1000\ \mathrm{km}\)、\(f=10^{-4}\ \mathrm{s^{-1}}\),求 \(Ro\);同一 flow 缩到 \(L=1\ \mathrm{km}\) 会怎样?
Answer 3
大尺度 \(Ro=10/[(10^{-4})(10^6)]=0.10\),rotation important;\(L=10^3\ \mathrm m\) 时 \(Ro=100\),inertia dominates,simple geostrophy 不适用。
- [K02] NH geostrophic current 向 east,low pressure 在哪一侧?若有 bottom friction,通常怎样跨越 contours?
Answer 4
low pressure 在左侧,即 north;friction weakens speed/Coriolis,pressure-gradient force 留下指向 low pressure 的 component,flow 斜向 low-pressure side。
- [K03] \(N=0.020\ \mathrm{s^{-1}}\)、\(k_h=m\),用 nonrotating dispersion 求 \(\omega\) 与 period。
Answer 5
\(\omega=N/\sqrt2=0.0141\ \mathrm{s^{-1}}\),period \(2\pi/\omega=444\ \mathrm s=7.40\ \mathrm{min}\);uniform Boussinesq assumptions 必须保留。
- [K03] ocean depth \(H=4000\ \mathrm m\),求 nondispersive shallow-water speed 和跨越 \(6000\ \mathrm{km}\) 的粗略时间。
Answer 6
\(c=\sqrt{(9.81)(4000)}=198\ \mathrm{m\,s^{-1}}\);\(t=6.0\times10^6/198=3.03\times10^4\ \mathrm s=8.42\ \mathrm h\)。真实 bathymetry、dispersion 与 coast 会改变 arrival。
- [K04] NH 中 eastward wind stress \(\tau_x=0.10\ \mathrm{Pa}\)、\(\rho_0=1025\ \mathrm{kg\,m^{-3}}\)、\(f=10^{-4}\ \mathrm{s^{-1}}\),求 Ekman transport direction and magnitude。
Answer 7
\(\boldsymbol\tau\times\hat{\mathbf k}\) 对 eastward \(\boldsymbol\tau\) 指 south;\(M_E=0.10/[(1025)(10^{-4})]=0.976\ \mathrm{m^2\,s^{-1}}\) southward,即 wind 右侧。
- [K04] 若 Example 4 的 mixed-layer depth 从 \(50\) m 加倍到 \(100\) m,其余相同,warming 如何变化?
Answer 8
\(\Delta T\propto1/h\),降为 \(0.317\ \mathrm K\);这是 fixed-depth slab result,entrainment 的 water temperature 可能进一步改变答案。
- [K05] 两个 ensemble 都有 20 members;A 的 spread 小但 truth 经常落在外面,B spread 大而观测 rank nearly uniform。哪一个更可靠?
Answer 9
B 更可能 calibrated/reliable;A underdispersive/overconfident。仍需比较 sharpness、bias、proper scores 与多案例 out-of-sample verification,不能只看一次 truth。
- [综合] 台风附近能否同时有 hydrostatic environment、gradient/geostrophic outer flow 与 nonhydrostatic convective cores?为何不矛盾?
Answer 10
可以。balance 由 scale/aspect ratio/\(Ro\) 决定:large-scale environment vertical 近 hydrostatic、outer circulation rotation important;kilometre-scale intense updrafts vertical acceleration and moist physics important。不同区域/尺度采用不同 asymptotic model,不是同一 parcel 同时被互相冲突的近似支配。
Exit ticket / 迁移检验
某海域 forecast 显示 sea-surface warming、surface wind 改变与降水 probability 上升。写出最少四项诊断,分别检验 heat budget、rotation balance、air–sea coupling 与 ensemble credibility。
Exit-ticket benchmark
heat:\(Q_{net}\)、mixed-layer depth、advection/entrainment ledger;rotation:\(Ro\) and pressure-gradient/Coriolis residual;coupling:wind stress/Ekman convergence、latent/sensible heat and SST response;forecast:initial-condition/data-assimilation increments、ensemble spread/reliability、model bias and hindcast skill。只有同时通过 budget and verification,才可把一致变化解释为可信 coupled mechanism。
References and source notes / 参考资料与使用说明
以下权威直接来源均在 2026-08-31 实际打开并核对。正文重新组织,全部图依据方程原创绘制;未复制来源插图。
- MIT OpenCourseWare, 12.333 Atmospheric and Ocean Circulations — Lecture Notes — hydrostatic/geostrophic balances、vorticity、waves、baroclinic processes 与 ocean circulation 的大学课程框架。
- NOAA Ocean Service, The Coriolis Effect — hemisphere-dependent deflection and large-scale current context。
- NOAA Ocean Exploration, What Causes Ocean Currents? — wind、density、tides、rotation 与 basin controls。
- NOAA Pacific Marine Environmental Laboratory, Air–Sea Fluxes — momentum、heat、moisture flux observations and bulk-flux context。
- NOAA/NCEP Environmental Modeling Center, Ensemble Forecasting Background — initial perturbations、ensemble spread and predictability rationale。
- NOAA/NCEP Environmental Modeling Center, Global Ensemble Forecast System — operational ensemble products and verification context。
Source-to-content map
| Source | Used for | Knowledge ID / Section | Model boundary or convention |
|---|---|---|---|
| MIT OCW 12.333 | hydrostatic/geostrophic dynamics, waves, instability, circulation | K01, K02, K03, K04 | undergraduate course hierarchy; selected topics extend toward Catalog [UG1→EXT] while derivations remain explicit |
| NOAA Coriolis | hemispheric deflection and current context | K02 | pedagogical explanation; \(Ro\) states scale boundary |
| NOAA Ocean Currents | multiple circulation drivers | K04 | avoids single-conveyor/single-force claim |
| NOAA PMEL Fluxes | air–sea momentum/heat/moisture exchanges | K04 | observed fluxes require turbulent bulk methods and uncertainty |
| NCEP Ensemble Background | perturbations, spread, predictability | K05 | ensemble spread requires calibration |
| NCEP GEFS | operational ensemble output/verification | K05 | products are model-cycle dependent, not universal probabilities |
复习与拓展
- Physical Principles in Living Systems 开始下一 Unit;本课的 transport、feedback、waves 与 uncertainty 会在 living systems 中以新的尺度重现。
- 可扩展:thermal wind、potential vorticity、Rossby/Kelvin waves、quasi-geostrophic theory、Sverdrup balance、baroclinic instability derivation、4D-Var/EnKF 与 forecast verification scores。
- 与 Climate and Planetary Energy Balance 的接口:weather resolves trajectory on days,climate studies distributions, budgets and forced response across longer scales;二者共享 coupled equations but ask different questions。