Question bank — Liouville's theorem — phase space volume conservation

False. $\rho$ sirf ek trajectory ke saath constant hoti hai ($d\rho /dt=0$); ek fixed jagah par $\partial \rho /\partial t$ bahut bada ho sakta hai jab koi dense blob andar ya bahar drift karta hai.

False. Volume (measure) conserved hai, lekin blob generally thin filaments mein stretch ho jaata hai jabki doosri direction mein patla ho jaata hai — yahi stretch-and-fold exactly hai jisse chaos aur mixing paida hoti hai.

False. Iske liye flow divergence-free honi chahiye, jo sirf Hamilton's equations ke through mixed-partial cancellation se guaranteed hai. Ek damped oscillator ($\dot{p}=\cdots -\gamma p$) ise violate karta hai.

False. Blob badalta zaroor hai — yeh ek rectangle se parallelogram mein shear ho jaata hai. Area isliye bachta hai kyunki shear perpendicular height ($\Delta p$, kyunki $\dot{p}=0$) ko untouched chhod deta hai.

False. Yeh origin ke around rigidly rotate karta hai (rescaled $Q,P$ mein circles). Rotation ka Jacobian determinant $1$ hota hai, isliye koi shrinking nahi hoti. Shrinking ke liye dissipation chahiye hogi.

False. Iska ulta: bounded, volume-preserving flow near-returns ko force karta hai — yahi Poincaré Recurrence Theorem hai. Conserved volume hi exactly recurrence ko unavoidable banata hai.

False. Ek single point ka volume trivially zero hota hai. Content ek cloud (ensemble) ke paas wale states ke saath flow karne ke baare mein hai — isliye yeh Statistical Mechanics — Ensembles ko underpin karta hai.

False. Sirf Canonical Transformations ise preserve karti hain (unka Jacobian symplectic hota hai, determinant $1$). Ek generic coordinate change volume ko rescale kar deta hai.

Error yeh hai: continuity kisi bhi conserved fluid ke liye hold karta hai, dissipative ho ya na ho. Liouville tab hi aata hai jab additionally prove karo ki $\nabla \cdot \dot{\mathbf{x}}=0$, jo $\rho (\nabla \cdot \dot{\mathbf{x}})$ term ko khatam kar deta hai.

Sign error: ${\dot{p}}_i=-\partial H/\partial q_i$, isliye doosra term $-{\partial }^{2}H/\partial p_i\partial q_i$ hai. Phir do mixed partials cancel hokar $0$ bante hain, add nahi hote.

Do errors hain. Friction $\nabla \cdot \dot{\mathbf{x}}=-\gamma \ne 0$ deta hai, isliye yeh Liouville system nahi hai, aur jab $\rho$ badhti hai toh theorem $d\rho /dt=0$ violated hota hai — points attractor par pile up ho jaate hain.

Error: $d\rho /dt$ material derivative hai (flow ke saath chalte hue). $\rho$ ab bhi space mein point-to-point vary karta hai aur ek fixed point par time mein vary kar sakta hai; yeh sirf ek phase point ke saath ride kar rahe observer ke liye constant hai.

Error: mixed-partial cancellation ${\partial }^{2}H/\partial q\partial p={\partial }^{2}H/\partial p\partial q$ $H(q,p,t)$ ke liye bhi hold karta hai. Flow phir bhi divergence-free hai; time-dependent $H$ bas $\partial \rho /\partial t\ne 0$ possible banata hai, jo theorem allow karta hai.

Error: $\partial \rho /\partial t=-\{\rho ,H\}$ rearrange hokar $d\rho /dt={\partial }_t\rho +\{\rho ,H\}=0$ banta hai. Nonzero bracket exactly ${\partial }_t\rho$ se balance hota hai; total derivative zero hota hai. Dekho Poisson Brackets.

Ek flowing blob ka volume rate $dV/dt=\int \nabla \cdot \dot{\mathbf{x}}dV$ se change hota hai. Agar integrand har jagah zero hai, toh integral zero hai, isliye $V$ frozen hai — wahi reason jisse incompressible fluids apna volume rakhte hain.

Kyunki hum exact state kabhi nahi jaante — bas yeh ki yeh kisi choti region mein kahin hai. Conserved volume ka matlab hai probability conserved hai, jo woh foundation hai jo Statistical Mechanics — Ensembles ko $\rho$ ko genuine probability fluid treat karne deta hai.

Yeh kehta hai Hamilton's Equations ki structure hi — ek $+\partial H/\partial p$ aur ek $-\partial H/\partial q$ shared $H$ ke saath — wahi hai jo $q$ aur $p$ expansions ko pair karke cancel karati hai. Yeh antisymmetric pairing Symplectic Geometry ka seed hai.

Shear layers ko ek axis ke parallel slide karta hai bina unhe perpendicular direction mein apart move kiye. Area = base × perpendicular height, aur height ($\Delta p$) untouched rehti hai, isliye area invariant rahta hai jabki shape skew ho jaati hai.

Liouville guarantee karta hai ki koi volume lose nahi hoti, trajectories ko crowd karne aur arbitrarily close aane ke liye force karta hai — yahi Poincaré Recurrence Theorem hai — lekin "close" "exact" nahi hai; measure conservation near-return deta hai, exact periodicity nahi.

Yeh $\rho$ ko incompressible conserved fluid treat karne ki permission deta hai, isliye ek stationary ensemble (${\partial }_t\rho =0$) ko $\{\rho ,H\}=0$ satisfy karna hoga — matlab sirf energy jaisi conserved quantities par depend karna hoga. Wahan se equilibrium ensembles aate hain.

Trivially uska zero volume zero rehta hai. Theorem ka real content sirf positive volume wale region ke liye kaam aata hai, matlab spread wale ensemble ke liye.

Dono nahi. Yeh exactly $\Delta q\Delta p$ rehta hai har time par; yeh bas ever-more-slanted, ever-thinner-looking parallelogram banta jaata hai jiska true area unchanged hai.

Haan. Divergence-free ek local statement hai har jagah, including fixed points; stillness mein baitha point koi expansion contribute nahi karta. Kisi bhi surrounding blob ka volume conservation unaffected hai.

Volume $e^{-\gamma t}\to 0$ ki tarah decay karta hai; blob origin attractor par collapse ho jaata hai. Yeh woh clean failure case hai jo prove karta hai ki Liouville genuinely Hamiltonian (divergence-free) structure require karta hai.

Haan. Divergence har instant par zero rehti hai (mixed partials cancel hote hain regardless of explicit $t$), isliye volume preserve hota hai tab bhi jab flow field khud time mein change ho raha ho.

Volume equality necessary hai lekin sufficient nahi. Flow ek specific volume-preserving (symplectic) map hai, isliye bahut saare equal-volume shapes ab bhi dynamics ke through ek doosre se unreachable hain.