With increasing globalization coupled with climate change, the spread of novel pests and pathogens in ecosystems is greater than ever before (Turner 2010, Raffa et al. 2008). These agents of disturbance have the potential to eliminate entire species and/or plant functional groups from forests (Atkins et al. 2020, Gough et al. in review). Yet, recent studies have indicated that forests have incredible capacity to maintain functional stability, including net primary production (NPP), despite significant disturbance-related losses in leaf area index (LAI) (Fahey et al. 2016, Grigri et al. 2020, Gough et al. 2020). Using a large-scale experiment at the University of Michigan Biological Station (UMBS) as a simulation of moderate to severe forest disturbance, the Forest Resilience Threshold Experiment (FoRTE) seeks to understand the mechanisms through which this stability is maintained in the face of such disturbances. Structures that remain after a disturbance may determine the stability of ecosystem processes following disturbance (Johnstone et al. 2016). Material legacies, or abiotic (e.g., nitrogen) and biotic (e.g., diversity) capital and structures (e.g., biomass) that persist after a disturbance are well studied in the context of plant community stability (Johnstone et al. 2016). Examples of such studies include analysis of: surviving tree species, abundance, and spatial arrangement (Seidl et al. 2014, Turner et al. 1998, Meigs and Keeton 2018); coarse downed and standing woody debris (Meigs and Keeton 2018); and soil seed bank, and seed persistence (Johnstone et al. 2016, Turner et al. 1998). Forest structural characteristics as functionally-relevant legacies of disturbance, however, have largely been excluded from prior studies at the ecosystem scale, despite growing evidence of their role in C cycling stability (Wales et al. 2020, Gough et al. 2020). Previous work at UMBS and elsewhere has established the relationship between structural characteristics and NPP in the absence of disturbance or immediately following disturbance, including measures of complexity such as rugosity (Hardiman et al. 2013b; Scheuerman et al. 2018; Gough et al. 2019), woody biomass (Gough et al. 2020; Haber et al. 2020) and, to a lesser degree, diversity (Gough et al. 2020; Gough et al. 2019; Gough et al. 2010; Scheuerman et al. 2018; Wales et al. 2020). In all of these cases, increased levels of each measure was tied to greater NPP. However, the question remains: how does the stability of multiple material legacies through disturbance affect the stability of ecosystem function? In this study, my overarching objective is to determine which of three categories of material legacies confer NPP stability, focusing on three-year resistance, as disturbance severity increases and in two disturbance types. Focusing on resistance, I will: 1.) Identify which structural and resource legacies with hypothesized linkages to NPP remain intact as disturbance severity increases and following two disturbance types; 2.) Evaluate how resource legacies relate to resistance as a dimension of NPP stability in two disturbance types, one affecting primarily large and the other small trees; 3.) Characterize which resource legacies convey NPP stability as disturbance severity increases from 0 to 85% gross LAI loss.