Sensory systems provide critical filters that enable organisms to detect and recognize valuable resources. Trophic cascades, structuring populations and communities, are determined to a large degree by trait-mediated interactions that rely on sensory inputs. Certain molecules serve as chemosensory stimuli and play keystone roles in determining outcomes of predator-prey dynamics at multiple trophic levels. Here, a surface-adsorbed glycoprotein signal molecule, MULTIFUNCin, is fully identified and its role in waveswept rocky intertidal community dynamics investigated. Barnacles (Balanus glandula), are constrained to produce MULTIFUNCin for cuticle/shell formation. This compound evoked settlement by conspecifics larvae in field assays, and thus, operates as seminal cue for recruitment. Moreover, the same substances triggered predation by a numerically dominant whelk species (Acanthinucella spirata) on barnacle juveniles and adults in lab and field experiments. This glycoprotein, therefore, balances simultaneous demographic processes that enhance, and diminish, barnacle populations. As dominant competitors for space, the relative equilibrium between barnacle recruitment and predation mortality has strong, cascading direct and indirect effects on community dynamics. Hence, MULTIFUNCin plays a keystone role within rocky intertidal habitats.

Chemical weathering influences many components of the Earth system, from nutrient supply to landscape evolution to long-term climate. Despite considerable advances in understanding what controls chemical weathering in theoretical models and laboratory experiments, there is still much uncertainty surrounding the controls of chemical weathering in nature. Here I present several studies on chemical erosion rates in steep, eroding terrain. In Chapter 2, I present a 1-D numerical model for the evolution of soil mineralogy on an eroding hillslope, to quantify how much fluctuations in physical erosion rates should affect soil composition and thus estimates of chemical erosion rates inferred from soil composition. In Chapter 3, I combine new measurements of mineral abundances in soil, saprolite, and bedrock with prior measurements of soil production rates, dust deposition rates, and chemical composition in soil, saprolite, bedrock, and dust at an intensely weathered site in Puerto Rico. These data suggest this suite of measurements can — for abundant, soluble mineral phases — produce estimates of long-term mineral-specific weathering rates with uncertainties smaller than 20% of the mean. Lastly, I discuss new measurements of soil production rates and rock and soil composition along two steep altitudinal (and hence climatic) transects in the Idaho Batholith. In Chapter 4, I show how these data may be combined with measurements of dust composition to quantify long-term dust deposition rates. Under the assumption that mafic-rich dust from the nearby Palouse loess has been mixed into the otherwise granitic Idaho soils, I calculate dust deposition rates of 3-13 t km^-2 yr^-1 at our field sites, consistent with modern dust deposition rates measured elsewhere in the western United States. In Chapter 5, I show that mean annual soil temperature exerts no discernible effect on chemical erosion rates or on the degree of chemical weathering across these Idaho field sites. These measurements also show that the degree of chemical weathering, but not the rate of chemical erosion, increases with (a) the annual duration of wet conditions in the soil, and (b) soil residence time. Contrary to many prior measurements in similar terrain, these measurements are consistent with kinetic-limited weathering, rather than supply-limited weathering.