Developing detailed landslide inventory maps of prehistoric landslides is essential to interpreting the frequency and conditions under which slopes have failed. When coupled with age estimates, landslide inventories can yield better predictions for future slope failures, thereby improving hazard assessment and increasing the chances for mitigation. Developing proxies for landslide age is an important area of research, however, age dating prehistoric landslides can be challenging due to sparse datable organic material within landslide deposits. In this study, surface roughness of the landslide deposit is used to construct a best fit age-roughness model that quantitatively assigns age based on smoothing of the deposit with time. A good place to test this model is in the Green River Watershed (GRW) area, located in King County, Washington. Hillslopes in this area are composed of glacial sediments are are prone to oversteepening caused by lateral migration of the Green River as well as failure from climatic and tectonic triggers. We examine the distribution of landslides in the GRW using high-resolution LIDAR data and show that the vast majority of landslides are downstream of a knickpoint which has created higher relief and more accommodation space for slope failure than upstream where the relief is lower with less accommodations space. Furthermore, we identify and map multiple river terraces below landslide deposits that can provide minimum age constraints for our age-roughness model. Different landslide age groups determined from an age-roughness model will allow us to determine the frequency of past landsliding in the GRW and the conditions under which they failed. By constraining the ages of these landslides we can determine if they failed during a known tectonic event or when climatic conditions were cooler and wetter. A greater number of landslide ages can be estimated from the age-roughness model than by traditional geochronology methods. This model will provide a means of increasing the landslide inventory database to increase our hazard assessment of future landslides in glaciolacustrine environments.