<?xml version='1.0'?>
<!DOCTYPE art SYSTEM 'http://www.biomedcentral.com/xml/article.dtd'>
<art>
   <ui>1687-2770-2011-138396</ui>
   <ji>1687-2770</ji>
   <fm>
      <dochead>Research Article</dochead>
      <bibl>
         <title>
            <p>Two-Dimension Riemann Initial-Boundary Value Problem of Scalar Conservation Laws with Curved Boundary</p>
         </title>
         <aug>
            <au id="A1"><snm>Chen</snm><fnm>Huazhou</fnm><insr iid="I1"/><insr iid="I2"/><email>chinkashyuu@yahoo.com.cn</email></au>
            <au ca="yes" id="A2"><snm>Pan</snm><fnm>Tao</fnm><insr iid="I2"/><email>tpan@jnu.edu.cn</email></au>
         </aug>
         <insg>
            <ins id="I1"><p>Department of Mathematics, Shanghai University, Shanghai 200444, China</p></ins>
            <ins id="I2"><p>Key Laboratory of Optoelectronic Information and Sensing Technologies of Guangdong Higher Educational Institutes, Jinan University, Guangzhou 510632, China</p></ins>
         </insg>
         <source>Boundary Value Problems</source>
         <issn>1687-2770</issn>
         <pubdate>2011</pubdate>
         <volume>2011</volume>
         <issue>1</issue>
         <fpage>138396</fpage>
         <url>http://www.boundaryvalueproblems.com/content/2011/1/138396</url>
         <xrefbib><pubid idtype="doi">10.1155/2011/138396</pubid></xrefbib>
      </bibl>
      <history><rec><date><day>16</day><month>12</month><year>2010</year></date></rec><acc><date><day>1</day><month>2</month><year>2011</year></date></acc><pub><date><day>24</day><month>2</month><year>2011</year></date></pub></history>
      <cpyrt><year>2011</year><collab>Huazhou Chen and Tao Pan.</collab><note>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.</note></cpyrt>
      <abs>
         <sec>
            <st>
               <p/>
            </st>
            <p>This paper is concerned with the structure of the weak entropy solutions to two-dimension Riemann initial-boundary value problem with curved boundary. Firstly, according to the definition of weak entropy solution in the sense of Bardos-Leroux-Nedelec (1979), the necessary and sufficient condition of the weak entropy solutions with piecewise smooth is given. The boundary entropy condition and its equivalent formula are proposed. Based on Riemann initial value problem, weak entropy solutions of Riemann initial-boundary value problem are constructed, the behaviors of solutions are clarified, and we focus on verifying that the solutions satisfy the boundary entropy condition. For different Riemann initial-boundary value data, there are a total of five different behaviors of weak entropy solutions. Finally, a worked-out specific example is given.</p>
         </sec>
      </abs>
   </fm>
   <bdy>
      <sec>
         <st>
            <p>1. Introduction</p>
         </st>
         <p>Multidimensional conservation laws are a famous hard problem that plays an important role in mechanics and physics [<abbr bid="B1">1</abbr>&#8211;<abbr bid="B3">3</abbr>]. For Cauchy problem of multi-dimensional scalar conservation laws, Conway and Smoller [<abbr bid="B4">4</abbr>] and Kruzkov [<abbr bid="B1">1</abbr>] have proved that weak solution uniquely exists if it also satisfies entropy condition, and it is called weak entropy solutions. In order to further understand qualitative behavior of solutions, it is also important to investigate multi-dimensional Riemann problems. For two-dimensional case, Lindquist [<abbr bid="B5">5</abbr>], Wagner [<abbr bid="B6">6</abbr>], Zhang and Zheng [<abbr bid="B7">7</abbr>] Guckenheimer [<abbr bid="B8">8</abbr>], Zheng [<abbr bid="B9">9</abbr>] among others, have discussed some relating Riemann problems. In a previous discussion, initial value contains several constant states with discontinuity lines so that self-similar transformations can be applied to reduce two-dimensional problem to one-dimensional case. The situation that initial value contains two constant states divided by a curve can not be solved by selfsimilar transformations, and Yang [<abbr bid="B10">10</abbr>] proposed a new approach for construction of shock wave and rarefaction wave solutions; especially, rarefaction wave was got by constructing implicit function instead of the usual selfsimilar method. This approach can be expanded to general <inline-formula><graphic file="1687-2770-2011-138396-i1.gif"/></inline-formula>-dimension. In addition, multi-dimensional scalar conservation laws with boundary are more common in practical problems. Bardos et al. [<abbr bid="B2">2</abbr>] have proved the existence and uniqueness of the weak entropy solution of initial-boundary problems of multi-dimensional scalar conservation laws. The main difficulty for nonlinear conservation laws with boundary is to have a good formation of the boundary condition. Namely, for a fixed initial value, we really can not impose such a condition at the boundary, and the boundary condition is necessarily linked to the entropy condition. Moreover the behavior of solutions for one-dimensional problem with boundary was discussed in [<abbr bid="B11">11</abbr>&#8211;<abbr bid="B18">18</abbr>]. However, for multi-dimensional problem with boundary, the behaviors of solutions are still hard to study.</p>
         <p>In this paper, two-dimensional case as an example of Yang's multi-dimensional Riemann problem [<abbr bid="B10">10</abbr>] is expanded to the case with boundary. Considering two-dimensional Riemann problem for scalar conservation laws with curved boundary, </p>
         <p>
            <display-formula id="M11">
               <graphic file="1687-2770-2011-138396-i2.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i3.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i4.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i5.gif"/></inline-formula> are both constants, <inline-formula><graphic file="1687-2770-2011-138396-i6.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i7.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i8.gif"/></inline-formula> is a smooth manifold and divides <inline-formula><graphic file="1687-2770-2011-138396-i9.gif"/></inline-formula> into two infinite parts, <inline-formula><graphic file="1687-2770-2011-138396-i10.gif"/></inline-formula>, and <inline-formula><graphic file="1687-2770-2011-138396-i11.gif"/></inline-formula> and denote <inline-formula><graphic file="1687-2770-2011-138396-i12.gif"/></inline-formula>.</p>
         <p>In Section 2, weak entropy solution of Riemann initial-boundary value problem (1.1) is defined, and the boundary entropy condition is discussed. In Section 3, weak entropy solutions of the corresponding Riemann initial value problem are expressed. In Section 4, using the weak entropy solutions of the corresponding Riemann initial value problem, we construct the weak entropy solutions of Riemann initial-boundary value problem, and prove that they satisfy the boundary entropy condition. The weak entropy solutions include a total of five different shock and rarefaction wave solutions based on different Riemann data. Finally, in Section 5, we give a worked-out specific example.</p>
      </sec>
      <sec>
         <st>
            <p>2. Preliminaries</p>
         </st>
         <p>According to the definition of the weak entropy solution and the boundary entropy condition to the general initial-boundary problems of multi-dimensional scalar conservation laws which was proposed by Bardos et al. [<abbr bid="B2">2</abbr>] and Pan and Lin [<abbr bid="B13">13</abbr>], we can obtain the following definition and three lemmas for Riemann initial-boundary value problem (1.1).</p>
         <p>Definition 2.1. </p>
         <p>A locally bounded and bounded variation function <inline-formula><graphic file="1687-2770-2011-138396-i13.gif"/></inline-formula> on <inline-formula><graphic file="1687-2770-2011-138396-i14.gif"/></inline-formula> is called a weak entropy solution of Riemann initial-boundary value problem (1.1) if, for any real constant <inline-formula><graphic file="1687-2770-2011-138396-i15.gif"/></inline-formula> and for any nonnegative function <inline-formula><graphic file="1687-2770-2011-138396-i16.gif"/></inline-formula>, it satisfies the following inequality: </p>
         <p>
            <display-formula id="M21">
               <graphic file="1687-2770-2011-138396-i17.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i18.gif"/></inline-formula> is the outward normal vector of curve <inline-formula><graphic file="1687-2770-2011-138396-i19.gif"/></inline-formula>.</p>
         <p>Lemma 2.2. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i20.gif"/></inline-formula> is a weak entropy solution of (1.1), then it satisfies the following boundary: entropy condition </p>
         <p>
            <display-formula id="M22">
               <graphic file="1687-2770-2011-138396-i21.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i22.gif"/></inline-formula>.</p>
         <p>It can be easily proved that <inline-formula><graphic file="1687-2770-2011-138396-i23.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i24.gif"/></inline-formula>, so (2.2) can be rewritten as </p>
         <p>
            <display-formula id="M23">
               <graphic file="1687-2770-2011-138396-i25.gif"/>
            </display-formula>
         </p>
         <p>thus one can get <inline-formula><graphic file="1687-2770-2011-138396-i26.gif"/></inline-formula> or </p>
         <p>
            <display-formula id="M24">
               <graphic file="1687-2770-2011-138396-i27.gif"/>
            </display-formula>
         </p>
         <p>and one notices that <inline-formula><graphic file="1687-2770-2011-138396-i28.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i29.gif"/></inline-formula>, then boundary entropy condition (2.2) is equivalent to </p>
         <p>
            <display-formula id="M25">
               <graphic file="1687-2770-2011-138396-i30.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>The proof for one-dimension case of Lemma 2.2 can be found in Pan and Lin's work [<abbr bid="B13">13</abbr>], and the proof for <inline-formula><graphic file="1687-2770-2011-138396-i31.gif"/></inline-formula>-dimension case is totally similar to one-dimension case; actually the idea of the proof first appears in Bardos et al.'s work [<abbr bid="B2">2</abbr>], so the proof details for Lemma 2.2 are omitted here.</p>
         <p>Lemma 2.3. </p>
         <p>A piecewise smooth function <inline-formula><graphic file="1687-2770-2011-138396-i32.gif"/></inline-formula> with smooth discontinuous surface is a weak entropy solution to the Riemann initial-boundary value problem (1.1) in the sense of (2.1) if and only if the following conditions are satisfied.</p>
         <p>(i) Rankine-Hugoniot condition: At any point <inline-formula><graphic file="1687-2770-2011-138396-i33.gif"/></inline-formula> on discontinuity surface <inline-formula><graphic file="1687-2770-2011-138396-i34.gif"/></inline-formula> of solution <inline-formula><graphic file="1687-2770-2011-138396-i35.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i36.gif"/></inline-formula> is a unit normal vector to <inline-formula><graphic file="1687-2770-2011-138396-i37.gif"/></inline-formula> at <inline-formula><graphic file="1687-2770-2011-138396-i38.gif"/></inline-formula> if </p>
         <p>
            <display-formula id="M26">
               <graphic file="1687-2770-2011-138396-i39.gif"/>
            </display-formula>
         </p>
         <p>then </p>
         <p>
            <display-formula id="M27">
               <graphic file="1687-2770-2011-138396-i40.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i41.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i42.gif"/></inline-formula>.</p>
         <p>For any constant <inline-formula><graphic file="1687-2770-2011-138396-i43.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i44.gif"/></inline-formula>, </p>
         <p>
            <display-formula id="M28">
               <graphic file="1687-2770-2011-138396-i45.gif"/>
            </display-formula>
         </p>
         <p>or equivalently </p>
         <p>
            <display-formula id="M29">
               <graphic file="1687-2770-2011-138396-i46.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>(ii) Boundary entropy condition: </p>
         <p>
            <display-formula id="M210">
               <graphic file="1687-2770-2011-138396-i47.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>(iii) Initial value condition: </p>
         <p>
            <display-formula id="M211">
               <graphic file="1687-2770-2011-138396-i48.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>For piecewise smooth solution with smooth discontinuous surface, Rankine-Hugoniot condition (2.7), entropy conditions (2.8), (2.9) and initial value condition (2.11) are obviously satisfied, see also the previous famous works in [<abbr bid="B4">4</abbr>, <abbr bid="B7">7</abbr>&#8211;<abbr bid="B9">9</abbr>]. As in Lemma 2.2, boundary entropy condition (2.10) also holds. The proof of the converse in not difficult and is omitted here.</p>
         <p>According to Bardos et al.'s work [<abbr bid="B2">2</abbr>], we have the following Lemma.</p>
         <p>Lemma 2.4. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i49.gif"/></inline-formula> is piecewise smooth weak entropy solution of (1.1) which satisfies the conditions of Lemma 2.3, then <inline-formula><graphic file="1687-2770-2011-138396-i50.gif"/></inline-formula> is unique.</p>
         <p>According to the uniqueness of weak entropy solution, as long as the piecewise smooth function satisfying Lemma 2.3 is constructed, the weak entropy solution of Riemann initial-boundary value problem can be obtained.</p>
      </sec>
      <sec>
         <st>
            <p>3. Solution of Riemann Initial Value Problem</p>
         </st>
         <p>First, we study the Riemann initial value problem corresponding to the Riemann initial-boundary value problem (1.1) as follows: </p>
         <p>
            <display-formula id="M31">
               <graphic file="1687-2770-2011-138396-i51.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>Condition <inline-formula><graphic file="1687-2770-2011-138396-i52.gif"/></inline-formula> For <inline-formula><graphic file="1687-2770-2011-138396-i53.gif"/></inline-formula>,&#8201; </p>
         <p>
            <display-formula id="M32">
               <graphic file="1687-2770-2011-138396-i54.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i55.gif"/></inline-formula> is a certain interval <inline-formula><graphic file="1687-2770-2011-138396-i56.gif"/></inline-formula> can be a finite number or <inline-formula><graphic file="1687-2770-2011-138396-i57.gif"/></inline-formula>.</p>
         <p>Condition <inline-formula><graphic file="1687-2770-2011-138396-i58.gif"/></inline-formula> combines flux functions <inline-formula><graphic file="1687-2770-2011-138396-i59.gif"/></inline-formula> and curved boundary manifold <inline-formula><graphic file="1687-2770-2011-138396-i60.gif"/></inline-formula>, providing necessary condition for the convex property of the new flux function which will be constructed in formula (4.5). The convex property clarifies whether the characteristics intersect or not, whether the weak solution satisfied internal entropy conditions (2.8) and (2.9) and boundary entropy condition (2.10), In addition, Condition <inline-formula><graphic file="1687-2770-2011-138396-i61.gif"/></inline-formula> is easily satisfied, for example, <inline-formula><graphic file="1687-2770-2011-138396-i62.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i63.gif"/></inline-formula>, then <inline-formula><graphic file="1687-2770-2011-138396-i64.gif"/></inline-formula>, so Condition <inline-formula><graphic file="1687-2770-2011-138396-i65.gif"/></inline-formula> holds. Here <inline-formula><graphic file="1687-2770-2011-138396-i66.gif"/></inline-formula> is a cubic curve on the <inline-formula><graphic file="1687-2770-2011-138396-i67.gif"/></inline-formula> plane, and it is strictly bending.</p>
         <p>Yang's work [<abbr bid="B10">10</abbr>] showed that depending on whether the characteristics intersect or not, the weak entropy solution of (3.1) has two forms as follows.</p>
         <p>Lemma 3.1 (see [<abbr bid="B10">10</abbr>]). </p>
         <p>Suppose (<inline-formula><graphic file="1687-2770-2011-138396-i68.gif"/></inline-formula>) holds. If <inline-formula><graphic file="1687-2770-2011-138396-i69.gif"/></inline-formula>, then weak entropy solution of (3.1) is shock wave solution <inline-formula><graphic file="1687-2770-2011-138396-i70.gif"/></inline-formula>, and </p>
         <p>
            <display-formula id="M33">
               <graphic file="1687-2770-2011-138396-i71.gif"/>
            </display-formula>
         </p>
         <p>and discontinuity surface <inline-formula><graphic file="1687-2770-2011-138396-i72.gif"/></inline-formula> is </p>
         <p>
            <display-formula id="M34">
               <graphic file="1687-2770-2011-138396-i73.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i74.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i75.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i76.gif"/></inline-formula>.</p>
         <p>Lemma 3.2 (see [<abbr bid="B10">10</abbr>]). </p>
         <p>Suppose that (<inline-formula><graphic file="1687-2770-2011-138396-i77.gif"/></inline-formula>) holds. If <inline-formula><graphic file="1687-2770-2011-138396-i78.gif"/></inline-formula>, then weak entropy solution of (3.1) is rarefaction wave solution <inline-formula><graphic file="1687-2770-2011-138396-i79.gif"/></inline-formula>, and </p>
         <p>
            <display-formula id="M35">
               <graphic file="1687-2770-2011-138396-i80.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i81.gif"/></inline-formula> is the implicit function which satisfies </p>
         <p>
            <display-formula id="M36">
               <graphic file="1687-2770-2011-138396-i82.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>Theorem 3.3 (see [<abbr bid="B10">10</abbr>]). </p>
         <p>Suppose that (<inline-formula><graphic file="1687-2770-2011-138396-i83.gif"/></inline-formula>) holds. Given <inline-formula><graphic file="1687-2770-2011-138396-i84.gif"/></inline-formula>, then</p>
         <p indent="1">(i)if <inline-formula><graphic file="1687-2770-2011-138396-i85.gif"/></inline-formula>, weak entropy solution of (3.1) is <inline-formula><graphic file="1687-2770-2011-138396-i86.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i87.gif"/></inline-formula> has a form as (3.3);</p>
         <p indent="1">(ii)if <inline-formula><graphic file="1687-2770-2011-138396-i88.gif"/></inline-formula>, weak entropy solution of (3.1) is <inline-formula><graphic file="1687-2770-2011-138396-i89.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i90.gif"/></inline-formula> has a form as (3.5);</p>
         <p indent="1">(iii)weak entropy solutions formed as (3.3) and (3.5) uniquely exist.</p>
         <p/>
         <p>The weak entropy solutions constructed here are piecewise smooth and satisfy conditions (i) and (iii) of Lemma 2.3.</p>
      </sec>
      <sec>
         <st>
            <p>4. Solution of Riemann Initial-Boundary Value Problem</p>
         </st>
         <p>Now we restrict the weak entropy solutions of the Riemann initial value problem (3.1) constructed in Section 3 in region <inline-formula><graphic file="1687-2770-2011-138396-i91.gif"/></inline-formula>, and they still satisfy conditions (i) and (iii) of Lemma 2.3. If they also satisfy the boundary entropy condition (ii) of Lemma 2.3, then they are the weak entropy solutions of Riemann initial-boundary value problem (1.1).</p>
         <p>Based on different Riemann data of <inline-formula><graphic file="1687-2770-2011-138396-i92.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i93.gif"/></inline-formula>, the weak entropy solutions of the Riemann initial value problem (3.1) have the following five different behaviors when restricted in region <inline-formula><graphic file="1687-2770-2011-138396-i94.gif"/></inline-formula>.</p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i95.gif"/></inline-formula>, the solution of (3.1) is shock wave <inline-formula><graphic file="1687-2770-2011-138396-i96.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M41">
               <graphic file="1687-2770-2011-138396-i97.gif"/>
            </display-formula>
         </p>
         <p/>
         <p><inline-formula><graphic file="1687-2770-2011-138396-i98.gif"/></inline-formula> is formed by moving <inline-formula><graphic file="1687-2770-2011-138396-i99.gif"/></inline-formula> along the direction of the vector <inline-formula><graphic file="1687-2770-2011-138396-i100.gif"/></inline-formula>, and the outward normal vector <inline-formula><graphic file="1687-2770-2011-138396-i101.gif"/></inline-formula> of curve <inline-formula><graphic file="1687-2770-2011-138396-i102.gif"/></inline-formula> is equal to <inline-formula><graphic file="1687-2770-2011-138396-i103.gif"/></inline-formula>. According to the angle between <inline-formula><graphic file="1687-2770-2011-138396-i104.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i105.gif"/></inline-formula>, the solution restricted in <inline-formula><graphic file="1687-2770-2011-138396-i106.gif"/></inline-formula> has two behaviors as follows.</p>
         <p>Case 1. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i107.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i108.gif"/></inline-formula>.</p>
         <p>See also Figure <figr fid="F1">1(a)</figr>; it shows that the angle between <inline-formula><graphic file="1687-2770-2011-138396-i109.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i110.gif"/></inline-formula> is an acute angle, the shock wave surface <inline-formula><graphic file="1687-2770-2011-138396-i111.gif"/></inline-formula> is outside region <inline-formula><graphic file="1687-2770-2011-138396-i112.gif"/></inline-formula>, and the solution is constant state formed as </p>
         <p>
            <display-formula id="M42">
               <graphic file="1687-2770-2011-138396-i113.gif"/>
            </display-formula>
         </p>
         <p/>
         <fig id="F1"><title><p>Figure 1</p></title><caption><p>Case 1.</p></caption><text>
   <p><b>Case 1.</b>The constant solutionThe phase plane <inline-formula><graphic file="1687-2770-2011-138396-i114.gif"/></inline-formula></p>
</text><graphic file="1687-2770-2011-138396-1"/></fig>
         <p>Case 2. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i115.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i116.gif"/></inline-formula>.</p>
         <p>See also Figure <figr fid="F2">2</figr>; it shows that the angle between <inline-formula><graphic file="1687-2770-2011-138396-i117.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i118.gif"/></inline-formula> is an obtuse angle, the shock wave surface <inline-formula><graphic file="1687-2770-2011-138396-i119.gif"/></inline-formula> is inside region <inline-formula><graphic file="1687-2770-2011-138396-i120.gif"/></inline-formula>, and the solution is shock wave formed as </p>
         <p>
            <display-formula id="M43">
               <graphic file="1687-2770-2011-138396-i121.gif"/>
            </display-formula>
         </p>
         <p/>
         <fig id="F2"><title><p>Figure 2</p></title><caption><p>The shock wave solution of Case 2.</p></caption><text>
   <p>
      <b>The shock wave solution of Case 2.</b>
   </p>
</text><graphic file="1687-2770-2011-138396-2"/></fig>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i122.gif"/></inline-formula>, the solution of (3.1) is rarefaction wave <inline-formula><graphic file="1687-2770-2011-138396-i123.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M44">
               <graphic file="1687-2770-2011-138396-i124.gif"/>
            </display-formula>
         </p>
         <p/>
         <p><inline-formula><graphic file="1687-2770-2011-138396-i125.gif"/></inline-formula> is formed by moving <inline-formula><graphic file="1687-2770-2011-138396-i126.gif"/></inline-formula> along the direction of the vector <inline-formula><graphic file="1687-2770-2011-138396-i127.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i128.gif"/></inline-formula> is formed by moving <inline-formula><graphic file="1687-2770-2011-138396-i129.gif"/></inline-formula> along the direction of the vector <inline-formula><graphic file="1687-2770-2011-138396-i130.gif"/></inline-formula>, and the outward normal vector <inline-formula><graphic file="1687-2770-2011-138396-i131.gif"/></inline-formula> of curve <inline-formula><graphic file="1687-2770-2011-138396-i132.gif"/></inline-formula> is equal to <inline-formula><graphic file="1687-2770-2011-138396-i133.gif"/></inline-formula>.</p>
         <p>We construct a new flux function </p>
         <p>
            <display-formula id="M45">
               <graphic file="1687-2770-2011-138396-i134.gif"/>
            </display-formula>
         </p>
         <p>according to condition (<inline-formula><graphic file="1687-2770-2011-138396-i135.gif"/></inline-formula>), <inline-formula><graphic file="1687-2770-2011-138396-i136.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i137.gif"/></inline-formula> is convex, and <inline-formula><graphic file="1687-2770-2011-138396-i138.gif"/></inline-formula> is monotonically increasing function, so <inline-formula><graphic file="1687-2770-2011-138396-i139.gif"/></inline-formula>. And also </p>
         <p>
            <display-formula id="M46">
               <graphic file="1687-2770-2011-138396-i140.gif"/>
            </display-formula>
         </p>
         <p>Thus, <inline-formula><graphic file="1687-2770-2011-138396-i141.gif"/></inline-formula>. According to the angles between <inline-formula><graphic file="1687-2770-2011-138396-i142.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i143.gif"/></inline-formula>, and <inline-formula><graphic file="1687-2770-2011-138396-i144.gif"/></inline-formula>, the solution restricted in <inline-formula><graphic file="1687-2770-2011-138396-i145.gif"/></inline-formula> has three behaviors as follows.</p>
         <p>Case 3. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i146.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i147.gif"/></inline-formula>.</p>
         <p>See also Figure <figr fid="F3">3</figr>; it shows that the angles between <inline-formula><graphic file="1687-2770-2011-138396-i148.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i149.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i150.gif"/></inline-formula> are obtuse angles, the rarefaction wave surfaces <inline-formula><graphic file="1687-2770-2011-138396-i151.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i152.gif"/></inline-formula> are both inside region <inline-formula><graphic file="1687-2770-2011-138396-i153.gif"/></inline-formula>, and the solution is rarefaction wave formed as </p>
         <p>
            <display-formula id="M47">
               <graphic file="1687-2770-2011-138396-i154.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i155.gif"/></inline-formula> is the implicit function which satisfies (3.6).</p>
         <fig id="F3"><title><p>Figure 3</p></title><caption><p>The rarefaction wave solution of Case 3.</p></caption><text>
   <p>
      <b>The rarefaction wave solution of Case 3.</b>
   </p>
</text><graphic file="1687-2770-2011-138396-3"/></fig>
         <p>Case 4. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i156.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i157.gif"/></inline-formula>.</p>
         <p>See also Figure <figr fid="F4">4(a)</figr>; it shows that the angle between <inline-formula><graphic file="1687-2770-2011-138396-i158.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i159.gif"/></inline-formula> is an obtuse angle, the angle between <inline-formula><graphic file="1687-2770-2011-138396-i160.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i161.gif"/></inline-formula> is an acute angles, the rarefaction wave surface <inline-formula><graphic file="1687-2770-2011-138396-i162.gif"/></inline-formula> is inside region <inline-formula><graphic file="1687-2770-2011-138396-i163.gif"/></inline-formula>, the rarefaction wave surface <inline-formula><graphic file="1687-2770-2011-138396-i164.gif"/></inline-formula> is outside region <inline-formula><graphic file="1687-2770-2011-138396-i165.gif"/></inline-formula>, and the solution is rarefaction wave formed as </p>
         <p>
            <display-formula id="M48">
               <graphic file="1687-2770-2011-138396-i166.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i167.gif"/></inline-formula> is the implicit function which satisfies (3.6).</p>
         <fig id="F4"><title><p>Figure 4</p></title><caption><p>Case 4.</p></caption><text>
   <p><b>Case 4.</b>The rarefaction wave solutionThe phase plane <inline-formula><graphic file="1687-2770-2011-138396-i168.gif"/></inline-formula></p>
</text><graphic file="1687-2770-2011-138396-4"/></fig>
         <p>Case 5. </p>
         <p>If <inline-formula><graphic file="1687-2770-2011-138396-i169.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i170.gif"/></inline-formula>.</p>
         <p>See also Figure <figr fid="F5">5(a)</figr>; it shows that the angles between <inline-formula><graphic file="1687-2770-2011-138396-i171.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i172.gif"/></inline-formula>, and <inline-formula><graphic file="1687-2770-2011-138396-i173.gif"/></inline-formula> are acute angles, the rarefaction wave surfaces <inline-formula><graphic file="1687-2770-2011-138396-i174.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i175.gif"/></inline-formula> are both outside region <inline-formula><graphic file="1687-2770-2011-138396-i176.gif"/></inline-formula>, and the solution is constant state formed as </p>
         <p>
            <display-formula id="M49">
               <graphic file="1687-2770-2011-138396-i177.gif"/>
            </display-formula>
         </p>
         <p/>
         <fig id="F5"><title><p>Figure 5</p></title><caption><p>Case 5.</p></caption><text>
   <p><b>Case 5.</b>The constant solutionThe phase plane <inline-formula><graphic file="1687-2770-2011-138396-i178.gif"/></inline-formula></p>
</text><graphic file="1687-2770-2011-138396-5"/></fig>
         <p>Next, we verify the above five solutions all satisfying the boundary entropy condition (ii) of Lemma 2.3. By noticing the definition of <inline-formula><graphic file="1687-2770-2011-138396-i179.gif"/></inline-formula> (4.5) and its convex property, the boundary entropy condition (ii) of Lemma 2.3 can be equivalent to the following formula </p>
         <p>
            <display-formula id="M410">
               <graphic file="1687-2770-2011-138396-i180.gif"/>
            </display-formula>
         </p>
         <p>and thus we verify the above five solutions all satisfying the boundary entropy condition (4.10).</p>
         <p>Case 1. </p>
         <p>When <inline-formula><graphic file="1687-2770-2011-138396-i181.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i182.gif"/></inline-formula>, the shock wave solution is formed as (4.2). In this case, <inline-formula><graphic file="1687-2770-2011-138396-i183.gif"/></inline-formula> since </p>
         <p>
            <display-formula id="M411">
               <graphic file="1687-2770-2011-138396-i184.gif"/>
            </display-formula>
         </p>
         <p>and <inline-formula><graphic file="1687-2770-2011-138396-i185.gif"/></inline-formula>, where <inline-formula><graphic file="1687-2770-2011-138396-i186.gif"/></inline-formula> is the extreme point of <inline-formula><graphic file="1687-2770-2011-138396-i187.gif"/></inline-formula>. For <inline-formula><graphic file="1687-2770-2011-138396-i188.gif"/></inline-formula>, according to the convex property of <inline-formula><graphic file="1687-2770-2011-138396-i189.gif"/></inline-formula>, we have that </p>
         <p>
            <display-formula id="M412">
               <graphic file="1687-2770-2011-138396-i190.gif"/>
            </display-formula>
         </p>
         <p>and so the boundary entropy condition (4.10) is verified.</p>
         <p>Case 2. </p>
         <p>When <inline-formula><graphic file="1687-2770-2011-138396-i191.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i192.gif"/></inline-formula>, the shock wave solution is formed as (4.3). In this case, <inline-formula><graphic file="1687-2770-2011-138396-i193.gif"/></inline-formula>, so the boundary entropy condition (4.10) is naturally verified.</p>
         <p>Case 3. </p>
         <p>When <inline-formula><graphic file="1687-2770-2011-138396-i194.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i195.gif"/></inline-formula>, the rarefaction wave solution is formed as (4.7). In this case, <inline-formula><graphic file="1687-2770-2011-138396-i196.gif"/></inline-formula>, and so the boundary entropy condition (4.10) is naturally verified.</p>
         <p>Case 4. </p>
         <p>When <inline-formula><graphic file="1687-2770-2011-138396-i197.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i198.gif"/></inline-formula>, the rarefaction wave solution is formed as (4.8). In this case, <inline-formula><graphic file="1687-2770-2011-138396-i199.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i200.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i201.gif"/></inline-formula> (see also Figure <figr fid="F4">4(b)</figr>), namely, <inline-formula><graphic file="1687-2770-2011-138396-i202.gif"/></inline-formula>. For <inline-formula><graphic file="1687-2770-2011-138396-i203.gif"/></inline-formula>, according to the convex property of <inline-formula><graphic file="1687-2770-2011-138396-i204.gif"/></inline-formula> and Lagrange mean value theorem, there exists <inline-formula><graphic file="1687-2770-2011-138396-i205.gif"/></inline-formula>, satisfying </p>
         <p>
            <display-formula id="M413">
               <graphic file="1687-2770-2011-138396-i206.gif"/>
            </display-formula>
         </p>
         <p>and so the boundary entropy condition (4.10) is verified.</p>
         <p>Case 5. </p>
         <p>When <inline-formula><graphic file="1687-2770-2011-138396-i207.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i208.gif"/></inline-formula>, the rarefaction wave solution is formed as (4.9). In this case, <inline-formula><graphic file="1687-2770-2011-138396-i209.gif"/></inline-formula> since <inline-formula><graphic file="1687-2770-2011-138396-i210.gif"/></inline-formula> (see also Figure <figr fid="F5">5(b)</figr>) For <inline-formula><graphic file="1687-2770-2011-138396-i211.gif"/></inline-formula>, according to the convex property of <inline-formula><graphic file="1687-2770-2011-138396-i212.gif"/></inline-formula> and Lagrange mean value theorem, there exists <inline-formula><graphic file="1687-2770-2011-138396-i213.gif"/></inline-formula>, satisfying </p>
         <p>
            <display-formula id="M414">
               <graphic file="1687-2770-2011-138396-i214.gif"/>
            </display-formula>
         </p>
         <p>and so the boundary entropy condition (4.10) is verified.</p>
         <p>In summary, we have the following theorem.</p>
         <p>Theorem 4.1. </p>
         <p>Suppose that (<inline-formula><graphic file="1687-2770-2011-138396-i215.gif"/></inline-formula>) holds. Given <inline-formula><graphic file="1687-2770-2011-138396-i216.gif"/></inline-formula>, then</p>
         <p indent="1">(i)if <inline-formula><graphic file="1687-2770-2011-138396-i217.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i218.gif"/></inline-formula>, the solution of (1.1) is constant state and has form as (4.2),</p>
         <p indent="1">(ii)if <inline-formula><graphic file="1687-2770-2011-138396-i219.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i220.gif"/></inline-formula>, the solution of (1.1) is shock wave <inline-formula><graphic file="1687-2770-2011-138396-i221.gif"/></inline-formula>, and has form as (4.3),</p>
         <p indent="1">(iii)if <inline-formula><graphic file="1687-2770-2011-138396-i222.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i223.gif"/></inline-formula>, the solution of (1.1) is rarefaction wave <inline-formula><graphic file="1687-2770-2011-138396-i224.gif"/></inline-formula> and has a form as (4.7),</p>
         <p indent="1">(iv)if <inline-formula><graphic file="1687-2770-2011-138396-i225.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i226.gif"/></inline-formula>, the solution of (1.1) is rarefaction wave <inline-formula><graphic file="1687-2770-2011-138396-i227.gif"/></inline-formula> and has a form as (4.8);</p>
         <p indent="1">(v)if <inline-formula><graphic file="1687-2770-2011-138396-i228.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i229.gif"/></inline-formula>, the solution of (1.1) is constant state and has a form as (4.9).</p>
         <p/>
         <p>In addition the solutions formed as (4.2), (4.3), (4.7), (4.8), and (4.9) uniquely exist.</p>
         <p>Corollary 4.2. </p>
         <p>Suppose that <inline-formula><graphic file="1687-2770-2011-138396-i230.gif"/></inline-formula> for <inline-formula><graphic file="1687-2770-2011-138396-i231.gif"/></inline-formula>. <inline-formula><graphic file="1687-2770-2011-138396-i232.gif"/></inline-formula> can be finite or <inline-formula><graphic file="1687-2770-2011-138396-i233.gif"/></inline-formula>, and when <inline-formula><graphic file="1687-2770-2011-138396-i234.gif"/></inline-formula>,</p>
         <p indent="1">(i)if <inline-formula><graphic file="1687-2770-2011-138396-i235.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i236.gif"/></inline-formula>, the solution of (1.1) is rarefaction wave <inline-formula><graphic file="1687-2770-2011-138396-i237.gif"/></inline-formula> and has a form as (4.7),</p>
         <p indent="1">(ii)if <inline-formula><graphic file="1687-2770-2011-138396-i238.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i239.gif"/></inline-formula>, the solution of (1.1) is rarefaction wave <inline-formula><graphic file="1687-2770-2011-138396-i240.gif"/></inline-formula> and has a form as (4.8), </p>
         <p indent="1">(iii)if <inline-formula><graphic file="1687-2770-2011-138396-i241.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i242.gif"/></inline-formula>, the solution of (1.1) is constant state and has a form as (4.9),</p>
         <p indent="1">(iv)if <inline-formula><graphic file="1687-2770-2011-138396-i243.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i244.gif"/></inline-formula>, the solution of (1.1) is constant state and has a form as (4.2),</p>
         <p indent="1">(v)if <inline-formula><graphic file="1687-2770-2011-138396-i245.gif"/></inline-formula> and&#8201;&#8201;<inline-formula><graphic file="1687-2770-2011-138396-i246.gif"/></inline-formula>, the solution of (1.1) is shock wave <inline-formula><graphic file="1687-2770-2011-138396-i247.gif"/></inline-formula> and has a form as (4.3).</p>
         <p/>
         <p>Corollary 4.3. </p>
         <p>The approach here for two-dimensional Riemann initial-boundary problem can be expanded to the case of general <inline-formula><graphic file="1687-2770-2011-138396-i248.gif"/></inline-formula>-dimension.</p>
      </sec>
      <sec>
         <st>
            <p>5. An Example</p>
         </st>
         <p>Solve the following Riemann initial-boundary problem: </p>
         <p>
            <display-formula id="M51">
               <graphic file="1687-2770-2011-138396-i249.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i250.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i251.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i252.gif"/></inline-formula>, and it denotes <inline-formula><graphic file="1687-2770-2011-138396-i253.gif"/></inline-formula>. Since <inline-formula><graphic file="1687-2770-2011-138396-i254.gif"/></inline-formula>, we easily get <inline-formula><graphic file="1687-2770-2011-138396-i255.gif"/></inline-formula>, and condition <inline-formula><graphic file="1687-2770-2011-138396-i256.gif"/></inline-formula> holds.</p>
         <p>According to the different data of <inline-formula><graphic file="1687-2770-2011-138396-i257.gif"/></inline-formula> and <inline-formula><graphic file="1687-2770-2011-138396-i258.gif"/></inline-formula>, the behavior of the solution to Riemann initial-boundary problem (5.1) has a total of five situations; they can be described by the following five cases: (i) <inline-formula><graphic file="1687-2770-2011-138396-i259.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i260.gif"/></inline-formula>; (ii) <inline-formula><graphic file="1687-2770-2011-138396-i261.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i262.gif"/></inline-formula>; (iii) <inline-formula><graphic file="1687-2770-2011-138396-i263.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i264.gif"/></inline-formula>; (iv) <inline-formula><graphic file="1687-2770-2011-138396-i265.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i266.gif"/></inline-formula>; (v) <inline-formula><graphic file="1687-2770-2011-138396-i267.gif"/></inline-formula>, <inline-formula><graphic file="1687-2770-2011-138396-i268.gif"/></inline-formula>.</p>
         <p>For case (i), <inline-formula><graphic file="1687-2770-2011-138396-i269.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M52">
               <graphic file="1687-2770-2011-138396-i270.gif"/>
            </display-formula>
         </p>
         <p>and thus the solution is constant state formed as </p>
         <p>
            <display-formula id="M53">
               <graphic file="1687-2770-2011-138396-i271.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>For case (ii), <inline-formula><graphic file="1687-2770-2011-138396-i272.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M54">
               <graphic file="1687-2770-2011-138396-i273.gif"/>
            </display-formula>
         </p>
         <p>and thus the solution is shock wave solution formed as </p>
         <p>
            <display-formula id="M55">
               <graphic file="1687-2770-2011-138396-i274.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>For case (iii), <inline-formula><graphic file="1687-2770-2011-138396-i275.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M56">
               <graphic file="1687-2770-2011-138396-i276.gif"/>
            </display-formula>
         </p>
         <p>namely, <inline-formula><graphic file="1687-2770-2011-138396-i277.gif"/></inline-formula>, thus the solution is rarefaction wave formed as </p>
         <p>
            <display-formula id="M57">
               <graphic file="1687-2770-2011-138396-i278.gif"/>
            </display-formula>
         </p>
         <p>Here, we only need to solve <inline-formula><graphic file="1687-2770-2011-138396-i279.gif"/></inline-formula>, where </p>
         <p>
            <display-formula id="M58">
               <graphic file="1687-2770-2011-138396-i280.gif"/>
            </display-formula>
         </p>
         <p>To solve the following equation of <inline-formula><graphic file="1687-2770-2011-138396-i281.gif"/></inline-formula>: </p>
         <p>
            <display-formula id="M59">
               <graphic file="1687-2770-2011-138396-i282.gif"/>
            </display-formula>
         </p>
         <p>using Cardano formula, we can get the unique solution as </p>
         <p>
            <display-formula id="M510">
               <graphic file="1687-2770-2011-138396-i283.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>Since <inline-formula><graphic file="1687-2770-2011-138396-i284.gif"/></inline-formula> is the solution of implicit function, we still need to verify <inline-formula><graphic file="1687-2770-2011-138396-i285.gif"/></inline-formula> satisfying the following three conditions: (a) <inline-formula><graphic file="1687-2770-2011-138396-i286.gif"/></inline-formula>; (b) <inline-formula><graphic file="1687-2770-2011-138396-i287.gif"/></inline-formula>; (c) <inline-formula><graphic file="1687-2770-2011-138396-i288.gif"/></inline-formula>. In fact, according to the next proposition, the above three conditions can be easily verified, and the detail the omitted here.</p>
         <p>Proposition 5.1. </p>
         <p>For any real number <inline-formula><graphic file="1687-2770-2011-138396-i289.gif"/></inline-formula>, the following formula holds: </p>
         <p>
            <display-formula id="M511">
               <graphic file="1687-2770-2011-138396-i290.gif"/>
            </display-formula>
         </p>
         <p/>
         <p>Proof. </p>
         <p>Let </p>
         <p>
            <display-formula id="M512">
               <graphic file="1687-2770-2011-138396-i291.gif"/>
            </display-formula>
         </p>
         <p>then <inline-formula><graphic file="1687-2770-2011-138396-i292.gif"/></inline-formula> satisfies </p>
         <p>
            <display-formula id="M513">
               <graphic file="1687-2770-2011-138396-i293.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i294.gif"/></inline-formula> must be one root of (5.13). In fact, <inline-formula><graphic file="1687-2770-2011-138396-i295.gif"/></inline-formula>. Equation (5.13) at most has one real root; but <inline-formula><graphic file="1687-2770-2011-138396-i296.gif"/></inline-formula> is its real root, thus <inline-formula><graphic file="1687-2770-2011-138396-i297.gif"/></inline-formula>, and the proposition holds.</p>
         <p>For case (iv), <inline-formula><graphic file="1687-2770-2011-138396-i298.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M514">
               <graphic file="1687-2770-2011-138396-i299.gif"/>
            </display-formula>
         </p>
         <p>namely, <inline-formula><graphic file="1687-2770-2011-138396-i300.gif"/></inline-formula>, and thus the solution is rarefaction wave formed as </p>
         <p>
            <display-formula id="M515">
               <graphic file="1687-2770-2011-138396-i301.gif"/>
            </display-formula>
         </p>
         <p>where <inline-formula><graphic file="1687-2770-2011-138396-i302.gif"/></inline-formula> has the same form as (5.10).</p>
         <p>For case (v), <inline-formula><graphic file="1687-2770-2011-138396-i303.gif"/></inline-formula> and </p>
         <p>
            <display-formula id="M516">
               <graphic file="1687-2770-2011-138396-i304.gif"/>
            </display-formula>
         </p>
         <p>namely, <inline-formula><graphic file="1687-2770-2011-138396-i305.gif"/></inline-formula>, and thus the solution is constant state formed as </p>
         <p>
            <display-formula id="M517">
               <graphic file="1687-2770-2011-138396-i306.gif"/>
            </display-formula>
         </p>
         <p/>
      </sec>
   </bdy>
   <bm>
      <ack>
         <sec>
            <st>
               <p>Acknowledgment</p>
            </st>
            <p>This work is supported by the National Natural Science Foundation of China (10771087, 61078040), the Natural Science Foundation of Guangdong Province (7005948).</p>
         </sec>
      </ack>
      <refgrp><bibl id="B1"><title><p>Generalized solutions of the Cauchy problem in the large for nonlinear equations of first order</p></title><aug><au><snm>Kruzkov</snm><fnm>SN</fnm></au></aug><source>Soviet Mathematics Doklady</source><pubdate>1969</pubdate><volume>10</volume><fpage>785</fpage><lpage>788</lpage></bibl><bibl id="B2"><title><p>First order quasilinear equations with boundary conditions</p></title><aug><au><snm>Bardos</snm><fnm>C</fnm></au><au><snm>le Roux</snm><fnm>AY</fnm></au><au><snm>N&#233;d&#233;lec</snm><fnm>J-C</fnm></au></aug><source>Communications in Partial Differential Equations</source><pubdate>1979</pubdate><volume>4</volume><issue>9</issue><fpage>1017</fpage><lpage>1034</lpage><xrefbib><pubid idtype="doi">10.1080/03605307908820117</pubid></xrefbib></bibl><bibl id="B3"><title><p>Weak solutions of a stochastic model for two-dimensional second grade fluids</p></title><aug><au><snm>Razafimandimby</snm><fnm>PA</fnm></au><au><snm>Sango</snm><fnm>M</fnm></au></aug><source>Boundary Value Problems</source><pubdate>2010</pubdate><volume>2010</volume><lpage>47</lpage></bibl><bibl id="B4"><title><p>Clobal solutions of the Cauchy problem for quasi-linear first-order equations in several space variables</p></title><aug><au><snm>Conway</snm><fnm>E</fnm></au><au><snm>Smoller</snm><fnm>J</fnm></au></aug><source>Communications on Pure and Applied Mathematics</source><pubdate>1966</pubdate><volume>19</volume><fpage>95</fpage><lpage>105</lpage></bibl><bibl id="B5"><title><p>The scalar Riemann problem in two spatial dimensions: piecewise smoothness of solutions and its breakdown</p></title><aug><au><snm>Lindquist</snm><fnm>WB</fnm></au></aug><source>SIAM Journal on Mathematical Analysis</source><pubdate>1986</pubdate><volume>17</volume><issue>5</issue><fpage>1178</fpage><lpage>1197</lpage><xrefbib><pubid idtype="doi">10.1137/0517082</pubid></xrefbib></bibl><bibl id="B6"><title><p>The Riemann problem in two space dimensions for a single conservation law</p></title><aug><au><snm>Wagner</snm><fnm>DH</fnm></au></aug><source>SIAM Journal on Mathematical Analysis</source><pubdate>1983</pubdate><volume>14</volume><issue>3</issue><fpage>534</fpage><lpage>559</lpage><xrefbib><pubid idtype="doi">10.1137/0514045</pubid></xrefbib></bibl><bibl id="B7"><title><p>Two-dimensional Riemann problem for a single conservation law</p></title><aug><au><snm>Zhang</snm><fnm>T</fnm></au><au><snm>Zheng</snm><fnm>YX</fnm></au></aug><source>Transactions of the American Mathematical Society</source><pubdate>1989</pubdate><volume>312</volume><issue>2</issue><fpage>589</fpage><lpage>619</lpage><xrefbib><pubid idtype="doi">10.1090/S0002-9947-1989-0930070-3</pubid></xrefbib></bibl><bibl id="B8"><title><p>Shocks and rarefactions in two space dimensions</p></title><aug><au><snm>Guckenheimer</snm><fnm>J</fnm></au></aug><source>Archive for Rational Mechanics and Analysis</source><pubdate>1975</pubdate><volume>59</volume><issue>3</issue><fpage>281</fpage><lpage>291</lpage></bibl><bibl id="B9"><aug><au><snm>Zheng</snm><fnm>Y</fnm></au></aug><source>Systems of Conservation Laws: Two-Dimensional Riemann Problems, Progress in Nonlinear Differential Equations and Their Applications</source><publisher>Birkh&#228;user, Boston, Mass, USA</publisher><pubdate>2001</pubdate><volume>38</volume><fpage>xvi+317</fpage></bibl><bibl id="B10"><title><p>Multi-dimensional Riemann problem of scalar conservation law</p></title><aug><au><snm>Yang</snm><fnm>X</fnm></au></aug><source>Acta Mathematica Scientia</source><pubdate>1999</pubdate><volume>19</volume><issue>2</issue><fpage>190</fpage><lpage>200</lpage></bibl><bibl id="B11"><title><p>Boundary conditions for nonlinear hyperbolic systems of conservation laws</p></title><aug><au><snm>Dubois</snm><fnm>F</fnm></au><au><snm>LeFloch</snm><fnm>P</fnm></au></aug><source>Journal of Differential Equations</source><pubdate>1988</pubdate><volume>71</volume><issue>1</issue><fpage>93</fpage><lpage>122</lpage><xrefbib><pubid idtype="doi">10.1016/0022-0396(88)90040-X</pubid></xrefbib></bibl><bibl id="B12"><title><p>Explicit formula for weighted scalar nonlinear hyperbolic conservation laws</p></title><aug><au><snm>LeFloch</snm><fnm>P</fnm></au><au><snm>N&#233;d&#233;lec</snm><fnm>J-C</fnm></au></aug><source>Transactions of the American Mathematical Society</source><pubdate>1988</pubdate><volume>308</volume><issue>2</issue><fpage>667</fpage><lpage>683</lpage></bibl><bibl id="B13"><title><p>The global solution of the scalar nonconvex conservation law with boundary condition</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Lin</snm><fnm>L</fnm></au></aug><source>Journal of Partial Differential Equations</source><pubdate>1995</pubdate><volume>8</volume><issue>4</issue><fpage>371</fpage><lpage>383</lpage></bibl><bibl id="B14"><title><p>The global solution of the scalar nonconvex conservation law with boundary condition. (Continuation)</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Lin</snm><fnm>L</fnm></au></aug><source>Journal of Partial Differential Equations</source><pubdate>1998</pubdate><volume>11</volume><issue>1</issue><fpage>1</fpage><lpage>8</lpage></bibl><bibl id="B15"><title><p>Asymptotic stability of the rarefaction wave of a one-dimensional model system for compressible viscous gas with boundary</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Liu</snm><fnm>H</fnm></au><au><snm>Nishihara</snm><fnm>K</fnm></au></aug><source>Japan Journal of Industrial and Applied Mathematics</source><pubdate>1999</pubdate><volume>16</volume><issue>3</issue><fpage>431</fpage><lpage>441</lpage><xrefbib><pubid idtype="doi">10.1007/BF03167367</pubid></xrefbib></bibl><bibl id="B16"><title><p>Asymptotic behaviors of the solutions to scalar viscous conservation laws on bounded interval corresponding to rarefaction waves</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Jiu</snm><fnm>Q</fnm></au></aug><source>Progress in Natural Science</source><pubdate>1999</pubdate><volume>9</volume><issue>12</issue><fpage>948</fpage><lpage>952</lpage></bibl><bibl id="B17"><title><p>Asymptotic behaviors of the solution to an initial-boundary value problem for scalar viscous conservation laws</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Liu</snm><fnm>H</fnm></au></aug><source>Applied Mathematics Letters</source><pubdate>2002</pubdate><volume>15</volume><issue>6</issue><fpage>727</fpage><lpage>734</lpage><xrefbib><pubid idtype="doi">10.1016/S0893-9659(02)00034-4</pubid></xrefbib></bibl><bibl id="B18"><title><p>Asymptotic behavior of a one-dimensional compressible viscous gas with free boundary</p></title><aug><au><snm>Pan</snm><fnm>T</fnm></au><au><snm>Liu</snm><fnm>H</fnm></au><au><snm>Nishihara</snm><fnm>K</fnm></au></aug><source>SIAM Journal on Mathematical Analysis</source><pubdate>2002</pubdate><volume>34</volume><issue>2</issue><fpage>273</fpage><lpage>291</lpage><xrefbib><pubid idtype="doi">10.1137/S0036141001385745</pubid></xrefbib></bibl></refgrp>
   </bm>
</art>