Moso　bamboo　(Phyllostachys　pubescens)　stands　as　a　pivotal　economic　bamboo　species　globally,　holding　substantial　potential　for　carbon　sequestration.　Accurate　estimation　of　aboveground　biomass　(AGB)　in　Moso　bamboo　forests　is　crucial　due　to　its　close　ties　with　the　ecosystem＇s　carbon　cycle.　Despite　the　maturation　of　monitoring　techniques　for　Pantana　phyllostachysae　Chao,　a　significant　pest　of　Moso　bamboo,　its　interplay　with　AGB　in　these　forests　remains　enigmatic.　This　study　addressed　this　gap　by　categorizing　P.　phyllostachysae＇s　impact　on　Moso　bamboo　forests　into　four　levels:　healthy,　mild　damage,　moderate　damage,　and　severe　damage.　By　scrutinizing　field　data,　we　delved　into　the　shifts　in　Moso　bamboo　leaf　biomass　under　P.　phyllostachysae　stress.　Leveraging　Sentinel-2A/B　imagery,　we　extracted　diverse　correlation　factors,　including　original　wave　bands,　vegetation　indices,　texture　attributes,　and　vegetation＇s　physical　and　chemical　parameters.　Subsequently,　machine　learning　algorithms-namely,　random　forest　(RF),　support　vector　machine　(SVM),　and　extreme　gradient　boosting　(XGB)　were　employed　to　achieve　remote　sensing　inversion　of　AGB　in　Moso　bamboo　forests,　accounting　for　the　presence　of　insect　pests.　We　analyzed　the　response　of　Moso　bamboo　biomass　sensitive　factors　and　to　further　clarify　the　changes　of　AGB　of　Moso　bamboo　forests　under　insect　pest　stress　at　the　remote　sensing　level　ultimately.　The　results　showed　that　(1)　the　degree　of　Moso　bamboo　leaf　biomass　damage　was　positively　related　to　the　damage　level,　which　gradually　increased　from　15.15　%　to　59.42　%;　(2)　the　RF　algorithm　excelled　in　estimating　Moso　bamboo　forest　AGB,　particularly　in　May,　and　inclusion　of　insect　pest　considerations　enhanced　AGB　estimation　accuracy;　(3)　among　the　four　factor　types,　Band　information　and　vegetation　indices　emerged　as　most　impactful,　and　Band5,　Band11,　Band12,　NDVI68a　and　MSAVI　were　selected　the　most　often;　(4)　at　the　remote　sensing　level,　AGB　in　Moso　bamboo　forests　significantly　varies　under　P.　phyllostachysae　stress.　Healthy　areas　demonstrate　an　AGB　of　66.9037　Mg　ha−1,　while　heavily　affected　regions　drop　to　52.6591　Mg　ha−1.　It　can　be　seen　that　combining　pest　factors　for　Moso　bamboo　biomass　estimation　solves　the　problem　of　rough　biomass　estimation,　and　this　study　provides　a　more　promising　method　for　forest　growth　monitoring.　©　2024