QUANTIFYING SUSPENDED SEDIMENT CONCENTRATION FROM SPACE: GLOBAL VARIABILITY AND TRENDS

Recognizing its importance to water quality, river ecosystem health, estuary and delta dynamics, and nutrient and contaminant transport, riparian stakeholders and government agencies have monitored suspended sediment concentrations (SSC) around the world for >100 years. However, these monitoring projects are expensive and logistically difficult, and acquire data only at the measurement location. We present a novel method for estimating SSC using Landsat satellite imagery. The method is calibrated using more than 35,000 concurrent measurements of SSC and satellite imaging at ~300 diverse stations in the Western Hemisphere (r²>0.6). Google Earth Engine servers are used to apply this model to four decades of satellite imagery across the globe. For the period 1984 – present, we calculate SSC for every pixel classified as river water in millions of cloud-free satellite images, dramatically extending the spatial and temporal coverage of SSC monitoring. The results allow us to construct timeseries for the past four decades and to map longitudinal, regional, and global variations in SSC. Globally, we observe a range in SSC from 10⁰ – >10⁴ mg/L, with low concentrations in rivers draining resistant cratons and highest concentrations in poorly consolidated sediments, particularly in arid regions. Our method enables a host of geomorphic applications; we quantify the influence of dams and other human activities on downstream reaches, the role of tributary inputs, and the interplay between suspended sediment concentrations and channel morphology. Our approach does not eliminate the need for suspended sediment monitoring, as it requires the area of interest to be cloud-free, has lower temporal resolution than is necessary to capture peak sediment discharges, and, in the absence of site specific data, has irreducible error related to variations in grain composition, water chemistry (e.g. chlorophyll concentration), and satellite image processing. However, our ability to make hundreds or thousands of measurements along every major river in the world allows us to address persistent questions regarding suspended sediment source and fate in fluvial systems.