A　two-dimensional　dual-porosity　model　for　coupled　rainwater　and　landfill　gas　transport　through　capillary　barrier　covers　(CBCs)　was　established.　The　proposed　model　was　partially　verified　using　published　data　from　a　laboratory　loess　column　test.　By　fitting　the　field-measured　data　using　the　proposed　model,　the　optimal　dual-porosity　parameters　of　the　in-situ　loess　of　a　loess/gravel　CBC　in　Northwest　China　were　determined　to　be　wf　=　0.1　(volume　fraction　of　macropore　system),　Kf/Km　=　150　(intrinsic　permeability　ratio　of　macropore　to　micropore　system)　and　Re　=　1　(reduction　factor　of　mass　exchange　at　the　interface　between　macropore　and　micropore　systems).　This　model　was　used　to　predict　the　performance　of　the　CBC　experiencing　a　15-day　rainfall　period　followed　by　a　30-day　no-rainfall　period.　Results　showed　that　the　presence　of　macropores　induced　an　earlier　occurrence　of　percolation　accompanied　with　a　higher　percolation　rate.　The　cumulative　percolation　and　CH4　emission　rate　derived　from　the　dual-porosity　model　were　about　50　%　and　one　order　of　magnitude　higher　than　those　predicted　by　the　single　-porosity　model,　respectively.　In　addition,　the　existence　of　gravel　layer　had　remarkable　influence　on　minimizing　water　percolation　and　CH4　emission,　but　these　were　practically　independent　of　its　thickness.　The　results　of　this　study　can　provide　an　effective　method,　crucial　parameters　and　guidelines　for　the　design　of　CBCs.