Understanding　the　relationship　between　human　mobility　and　land　use　has　been　a　longstanding　topic　in　multiple　disciplines,　including　transport　geography　and　urban　planning.　Recently,　urban　collective　mobility　patterns　have　become　a　hot　research　direction　and　has　been　explored　at　an　unprecedented　space–time　scale　due　to　the　emerging　big　human　tracking　datasets　(e.g.,　mobile　phone　data).　However,　only　a　few　studies　have　comprehensively　quantified　the　effects　of　land　use　on　human　mobility　patterns　while　considering　the　influence　of　the　scale　of　spatial　analysis　units.　This　study　attempts　to　reinforce　this　knowledge　by　investigating　urban　human　convergence–divergence　patterns　and　their　relationship　with　land　use　distribution　characteristics　at　three　popular　types　of　spatial　analysis　units　of　human　mobility　studies　(voronoi　polygons,　grid　cells,　and　traffic　analysis　zones)　using　mobile　phone　data.　A　case　study　on　Shenzhen,　China　is　implemented,　and　results　indicate　that　eight　distinct　convergence–divergence　patterns　could　be　extracted　to　describe　urban　collective　mobility　patterns　despite　the　use　of　different　types　of　spatial　analysis　unit.　Moreover,　the　scale　of　spatial　analysis　units　exerts　a　few　effects　on　the　quantification　of　the　influence　of　land　use　distribution　on　human　convergence–divergence　patterns,　but　some　common　characteristics　could　be　summarized　from　these　discrepant　results.　The　findings　can　help　policy　makers　understand　urban　human　mobility　and　can　serve　as　a　guide　for　urban　management　and　planning.　©　2019