Skip to content


  • Research Article
  • Open Access

WKB Estimates for Linear Dynamic Systems on Time Scales

Advances in Difference Equations20082008:712913

  • Received: 3 May 2008
  • Accepted: 26 August 2008
  • Published:


We establish WKB estimates for linear dynamic systems with a small parameter on a time scale unifying continuous and discrete WKB method. We introduce an adiabatic invariant for dynamic system on a time scale, which is a generalization of adiabatic invariant of Lorentz_s pendulum. As an application we prove that the change of adiabatic invariant is vanishing as approaches zero. This result was known before only for a continuous time scale. We show that it is true for the discrete scale only for the appropriate choice of graininess depending on a parameter . The proof is based on the truncation of WKB series and WKB estimates.


  • Direct Calculation
  • Vector Function
  • Recurrence Relation
  • Matrix Function
  • Fundamental Matrix

1. Adiabatic Invariant of Dynamic Systems on Time Scales

Consider the following system with a small parameter on a time scale:
where is the delta derivative, is a -vector function, and

WKB method [1, 2] is a powerful method of the description of behavior of solutions of (1.1) by using asymptotic expansions. It was developed by Carlini (1817), Liouville, Green (1837) and became very useful in the development of quantum mechanics in 1920 [1, 3]. The discrete WKB approximation was introduced and developed in [48].

The calculus of times scales was initiated by Aulbach and Hilger [911] to unify the discrete and continuous analysis.

In this paper, we are developing WKB approximations for the linear dynamic systems on a time scale to unify the discrete and continuous WKB theory. Our formulas for WKB series are based on the representation of fundamental solutions of dynamic system (1.1) given in [12]. Note that the WKB estimate (see (2.21) below) has double asymptotical character and it shows that the error could be made small by either or

It is well known [13, 14] that the change of adiabatic invariant of harmonic oscillator is vanishing with the exponential speed as approaches zero, if the frequency is an analytic function.

In this paper, we prove that for the discrete harmonic oscillator (even for a harmonic oscillator on a time scale) the change of adiabatic invariant approaches zero with the power speed when the graininess depends on a parameter in a special way.

A time scale is an arbitrary nonempty closed subset of the real numbers. If has a left-scattered minimum , then otherwise Here we consider the time scales with and

For , we define forward jump operator
The forward graininess function is defined by

If , we say that is right scattered. If and , then is called right dense.

For and define the delta (see [10, 11]) derivative to be the number (provided it exists) with the property that for given any there exist a and a neighborhood of such that

for all .

For any positive define auxilliary "slow" time scales
with forward jump operator and graininess function

Further frequently we are suppressing dependence on or . To distinguish the differentiation by or we show the argument of differentiation in parenthesizes: or

Assuming (see [10] for the definition of rd-differentiable function), denote
where are unknown phase functions, is the Euclidean matrix norm, and are the exponential functions on a time scale [10, 11]:
Using the ratio of Wronskians formula proposed in [15] we introduce a new definition of adiabatic invariant of system (1.1)

Theorem 1.1.

Assume and for some positive number and any natural number conditions
are satisfied, where the positive parameter is so small that
Then for any solution of (1.1) and for all , the estimate

is true for some positive constant

Checking condition (1.16) of Theorem 1.1 is based on the construction of asymptotic solutions in the form of WKB series
where and
Here the functions are defined as
where is defined in (1.8), and are defined by recurrence relations

is the Kroneker symbol ( if , and otherwise).

In the next Theorem 1.2 by truncating series (1.20):

where are given in (1.21) and (1.22), we deduce estimate (1.16) from condition (1.26) below given directly in the terms of matrix

Theorem 1.2.

Assume that and conditions (1.14), (1.15), (1.17), and

are satisfied. Then, estimate (1.18) is true.

Note that if , then formulas (1.21) and (1.22) are simplified:
where from (1.8)
Taking in (1.25) and as in (1.21), we have
which means that in (1.20) and from (1.24)

Example 1.3.

Consider system (1.1) with Then for continuous time scale we have and by picking in (1.25) we get by direct calculations and
In view of
condition (1.26) under the assumption turns to

and from Theorem 1.2 we have the following corollary.

Corollary 1.4.

Assume that and (1.33) is satisfied. Then for estimate (1.18) with is true for all solutions of system (1.1) on continuous time scale

If , then (1.33) turns to

and for it is satisfied for any real .

If is an analytic function, then it is known (see [13]) that the change of adiabatic invariant approaches zero with exponential speed as approaches zero.

Example 1.5.

Consider harmonic oscillator on a discrete time scale ,
which could be written in form (1.1), where
Choosing from formulas (1.27) and (1.29) we have and
From (1.13) we get
If we choose

then all conditions of Theorem 1.2 are satisfied (see proof of Example 1.5 in the next section) for any real numbers , and estimate (1.18) with is true.

Note that for continuous time scale we have and (1.39) turns to the formula of adiabatic invariant for Lorentz's pendulum ([13]):

2. WKB Series and WKB Estimates

Fundamental system of solutions of (1.1) could be represented in form

where is an approximate fundamental matrix function and is an error vector function.

Introduce the matrix function

In [16], the following theory was proved.

Theorem 2.1.

Assume there exists a matrix function such that the matrix function is invertible, and the following exponential function on a time scale is bounded:
Then every solution of (1.1) can be represented in form (2.1) and the error vector function can be estimated as

where is the Euclidean vector (or matrix) norm.

Remark 2.2.

If , then from (2.4) we get


Indeed if , the function is increasing, so and from we get
and by integration
Note that from the definition
If , then the fundamental matrix in (2.1) is given by (see [12])

Lemma 2.3.

If conditions (1.14), (1.15) are satisfied, then

where the functions are defined in (1.9), (1.11).


By direct calculations (see [12]), we get from (2.11)
Using (2.14), we get
and from (1.14)
So by using (1.9), we have
From (2.2), (2.13), (2.17), we get (2.12) in view of

Proof of Theorem 1.1.

From (1.16) changing variable of integration we get
So using (2.12), we get
From this estimate and (2.5), we have

where is so small that (1.17) is satisfied. The last estimate follows from the inequality Indeed because is increasing for we have

Further from (2.1), (2.11), we have
Solving these equation for , we get
By multiplication (see (1.12)), we get
and using estimate (2.21), we have

Proof of Theorem 1.2.

Let us look for solutions of (1.1) in the form

where is given by (2.11), and functions are given via WKB series (1.20).

Substituting series (1.20) in (1.9), we get
To make asymptotically equal zero or we must solve for the equations
By direct calculations from the first quadratic equation
and the second one
we get two solutions given by (1.21) and (1.22). Note that
Furthermore from th equation

we get recurrence relations (1.22).

In view of Theorem 1.1, to prove Theorem 1.2 it is enough to deduce condition (1.16) from (1.26). By truncation of series (1.20) or by taking
we get (1.25). Defining as in (1.21) and (1.22), we have
Now (1.16) follows from (1.26) in view of
Note that from (1.13) and the estimates
it follows


From (1.37), (1.41), we have
and using (2.39), we get
Further for

So if , then (1.26) and all other conditions of Theorem 1.2 are satisfied, and (1.18) is true with .



The author wants to thank Professor Ondrej Dosly for his comments that helped improving the original manuscript.

Authors’ Affiliations

Kent State University, Stark Campus, 6000 Frank Avenue NW, Canton, OH 44720-7599, USA


  1. Fröman M, Fröman PO: JWKB-Approximation. Contributions to the Theory. North-Holland, Amsterdam, The Netherlands; 1965:viii+138.MATHGoogle Scholar
  2. Holmes MH: Introduction to Perturbation Methods, Texts in Applied Mathematics. Volume 20. Springer, New York, NY, USA; 1995:ix+337.View ArticleGoogle Scholar
  3. Birkhoff GD: Quantum mechanics and asymptotic series. Bulletin of the American Mathematical Society 1933, 39(10):681-700. 10.1090/S0002-9904-1933-05716-6MathSciNetView ArticleGoogle Scholar
  4. Braun PA: WKB method for three-term recursion relations and quasienergies of an anharmonic oscillator. Theoretical and Mathematical Physics 1979, 37(3):1070-1081.View ArticleGoogle Scholar
  5. Costin O, Costin R: Rigorous WKB for finite-order linear recurrence relations with smooth coefficients. SIAM Journal on Mathematical Analysis 1996, 27(1):110-134. 10.1137/S0036141093248037MATHMathSciNetView ArticleGoogle Scholar
  6. Dingle RB, Morgan GJ: methods for difference equations—I. Applied Scientific Research 1967, 18: 221-237.MATHMathSciNetView ArticleGoogle Scholar
  7. Geronimo JS, Smith DT: WKB (Liouville-Green) analysis of second order difference equations and applications. Journal of Approximation Theory 1992, 69(3):269-301. 10.1016/0021-9045(92)90003-7MATHMathSciNetView ArticleGoogle Scholar
  8. Wilmott P: A note on the WKB method for difference equations. IMA Journal of Applied Mathematics 1985, 34(3):295-302. 10.1093/imamat/34.3.295MATHMathSciNetView ArticleGoogle Scholar
  9. Aulbach B, Hilger S: Linear dynamic processes with inhomogeneous time scale. In Nonlinear Dynamics and Quantum Dynamical Systems (Gaussig, 1990), Mathematical Research. Volume 59. Akademie, Berlin, Germany; 1990:9-20.Google Scholar
  10. Bohner M, Peterson A: Dynamic Equations on Time Scales: An Introduction with Application. Birkhäuser, Boston, Mass, USA; 2001:x+358.View ArticleGoogle Scholar
  11. Hilger S: Analysis on measure chains—a unified approach to continuous and discrete calculus. Results in Mathematics 1990, 18(1-2):18-56.MATHMathSciNetView ArticleGoogle Scholar
  12. Hovhannisyan G:Asymptotic stability for linear dynamic systems on time scales. International Journal of Difference Equations 2007, 2(1):105-121.MathSciNetGoogle Scholar
  13. Littlewood JE: Lorentz's pendulum problem. Annals of Physics 1963, 21(2):233-242. 10.1016/0003-4916(63)90107-6MATHMathSciNetView ArticleGoogle Scholar
  14. Wasow W: Adiabatic invariance of a simple oscillator. SIAM Journal on Mathematical Analysis 1973, 4(1):78-88. 10.1137/0504009MATHMathSciNetView ArticleGoogle Scholar
  15. Hovhannisyan G, Taroyan Y:Adiabatic invariant for -connected linear oscillators. Journal of Contemporary Mathematical Analysis 1997, 31(6):47-57.MathSciNetGoogle Scholar
  16. Hovhannisyan G: Error estimates for asymptotic solutions of dynamic equations on time scales. In Proceedings of the 6th Mississippi State–UBA Conference on Differential Equations and Computational Simulations, 2007, San Marcos, Tex, USA, Electronic Journal of Differential Equations Conference. Volume 15. Southwest Texas State University; 159-162.Google Scholar


© Gro Hovhannisyan. 2008

This article is published under license to BioMed Central Ltd. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.