Understanding fracture distribution and its relation to knickpoint evolution in the Rio Icacos watershed (Luquillo Critical Zone Observatory, Puerto Rico) using landscape‐scale hydrogeophysics

Abstract

AbstractThe Rio Icacos watershed in the Luquillo Mountains (Puerto Rico) is unique due to its extremely rapid weathering rates. The watershed is incised into a quartz diorite that has developed a large knickzone defining the river profile. Regolith thickness within the watershed generally decreases from 20 to 30 m at the ridges to several meters in the quartz diorite‐dominated valley to tens of centimeters near the major river knickpoint, as determined from previous studies. Above the knickzone, we observe spheroidal corestones, but below this weathering is much less apparent. Measured erosion rates from previous studies are also high in the knickzone compared with upper elevations within the river profile. A suite of near‐surface geophysical methods (i.e. ground penetrating radar and terrain conductivity) capable of fast data acquisition in rugged landscapes, was deployed at kilometer scales to characterize critical zone structure. Concentrations of chaotic ground penetrating radar (GPR) reflections and diffraction hyperbolas with low electrical conductivity were observed in vertical zones that outcrop at the land surface as areas of intense fracturing and spheroidally weathered corestones. The width of these fractured and weathered zones showed an increase with proximity to the knickpoint, and was attributed to dilation of these sub‐vertical fractures near the knickpoint, as postulated theoretically by a stress model calculated for the topographic variability across the knickzone in the Rio Icacos, and that shows a release of compressive stress near the knickpoint. We hypothesize that erosion rates increase in the knickzone because of this inferred dilation of fractures. Specifically, opened fractures could enhance access of water and in turn promote spalling, erosion, and spheroidal weathering. This study shows that ground‐based hydrogeophysical methods used at the landscape‐scale (traditionally applied at smaller scales) can be used to explore critical zone architecture at the scales needed to explain the extreme variability in erosion rates across river profiles. © 2018 John Wiley & Sons, Ltd.