Correlating field and laboratory rates of particle abrasion, Rio Medio, Sangre de Cristo Mountains, New Mexico

River bed sediments commonly fine downstream due to a combination of particle abrasion, selective transport of finer grains, and fining of the local sediment supply from hillslopes and tributaries. Particle abrasion rates can be directly measured in the laboratory using tumbling barrels and annular flumes, however, scaling experimental particle abrasion rates to the field has proven difficult due to the confounding effects of selective transport and local supply variations. Here we attempt to correlate laboratory and field rates of particle abrasion in a field setting where these confounding effects can be controlled. The Rio Medio, which flows westward from the crest of the Sangre de Cristo Mountains in north central New Mexico, is one of several streams studied by John P. Miller in the early 1960's. Several kilometers downstream of its headwaters, the river crosses the Picuris-Pecos fault. Upstream of the fault the river receives quartzite, sandstone and shale clasts from the Ortega Formation, while downstream sediments are supplied by the Embudo Granite. Because the upstream lithologies are not resupplied downstream of the fault, any observed fining of these clasts should be due only to abrasion and selective transport. We hypothesize that we can account for the effects of selective transport by comparing relative fining rates for the different upstream lithologies from both the field and a laboratory tumbler. By correlating laboratory abrasion rates with rock strength, we can predict the relative fining rates due solely to abrasion expected in the field; differences between the predicted and observed fining rates could then be attributed to selective transport. We used point counts to measure bed surface sediment grain size distributions at 15 locations along a 25 kilometer reach of the Rio Medio, beginning just downstream of the fault and ending upstream of a developed area with disturbed channel conditions. We recorded intermediate particle diameter as well as lithologic composition for 100 clasts at each location. To better characterize the size distribution of poorly represented lithologies we also measured every grain we could find of these minority lithologies within a one square meter area on adjacent bar top surfaces. At each sampling site we also measured channel gradient, and bank-full width and depth. We collected gravel samples for laboratory tumbling experiments and larger bedrock blocks from which we extracted cores for the Brazilian tensile splitting strength test. Preliminary results show very rapid fining of the weak sedimentary rocks downstream of the fault, much less rapid fining of the quartzite and a net downstream coarsening of the granitic sediments, which dominate the bed in the downstream end of the study reach. This enigmatic downstream coarsening may be a legacy of Pliestocene glaciation, which is evident in the landscape upstream of the fault. Outburst floods or debris flows from upstream moraines may have delivered large quantities of coarse sediments to downstream reaches, which are now relatively immobile. Despite these complications, the Rio Medio site may yet provide sufficient information to test our proposed method for scaling laboratory particle abrasion rates to the field.