Abstract Although the importance of dynamic water storage and flowpath partitioning on discharge behavior has been well recognized within the critical zone community, there is still little consensus surrounding the question, “ How do climate factors from above and land characteristics from below dictate dynamic storage, flowpath partitioning, and ultimately regulate hydrological dynamics?” Answers to this question have been hindered by limited and inconsistent spatio-temporal data and arduous-to-measure subsurface data. Here we aim to answer this question above by using a semi-distributed hydrological model (HBV model) to simulate and understand the dynamics of water storage, groundwater flowpaths, and discharge in 15 headwater catchments across the contiguous United States. Results show that topography, precipitation falling as snow, and catchment soil texture all influence catchment dynamic storage, storage-discharge sensitivity, flowpath partitioning, and discharge flashiness. Flat, rain-dominated sites (< 30% precipitation as snow) with finer soils exhibited flashier discharge regimes than catchments with coarse soils and/or significant snowfall (>30% precipitation as snow). Rain-dominated sites with clay soils (indicative of chemical weathering) showed lower dynamic storage and discharge that was more sensitive to changes in dynamic storage than rainy sites with coarse soils. Steep, snowy sites with coarse soils (more mechanical weathering) had lowest dynamic storage and deep groundwater fed discharge that was less sensitive to changes in dynamic storage than fine-soil snowy or rainy catchments. These results highlight aridity and precipitation (snow versus rain) as the dominant climate controls from above and topography and soil texture as the dominant land controls from below. The study challenges the traditional view that climate controls water balance while subsurface structure dictates subsurface flow path. Rather, it shows that climate and land characteristics jointly regulate water balance, groundwater flowpath partitioning, and discharge responses. These findings have important implication for the projection of the future of water resources, especially as climate change and human activities continue to intensify.

Acid rain has degraded environmental health since the Industrial Revolution. Legislative efforts have 19 successfully reduced deposition rates, but recovery of affected ecosystems in the Mississippi River Basin 20 (MRB) remains to be assessed. Combining analysis of temporal trends of indicative chemical species in 21 streams and rivers with a Bayesian statistical model, we found strong evidence of reduced effect by acid 22 rain on water chemistry; however, the effect by agricultural activities and climate change are intensifying. 23 pH increased in the Eastern MRB, the historically more heavily affected region, and SO4 loading 24 decreased everywhere, suggesting recovery. Widespread fertilizer use, however, has likely accelerated 25 carbonate weathering and water acidification. As a result, water became more acidic in western sites and 26 annual divalent cation (DIV) load increased at all sites, showing statistically significant trends. Extended 27 dry summers under climate change have likely contributed to SO4 and DIV export via shale weathering in 28 the basin, when the groundwater table drops. We also found evidence that, while not a significant 29 contributor yet, ever increasing atmospheric CO2 levels will likely add to cation export from the MRB in 30 the future. Using long-term data over a large spatial scale, this study represents a comprehensive 31 assessment of the recovery of water chemistry in river and stream ecosystems from acid rain in the 32 Mississippi Basin, taking into consideration the entangled effects of agricultural activities, acid mine 33 drainage, and extended droughts and elevated atmospheric CO2 concentration under anthropogenic 34 climate change.

Excessive nutrients have penetrated water bodies worldwide yet our forecasting capabilities remain elusive. This work tests the shallow and deep hypothesis: chemical contrasts in the subsurface shape nitrate export patterns. We use data synthesis for 228 U.S. watersheds and reactive transport modeling (500 simulations) spanning broad climate, geology, and land conditions. Data synthesis showed that human perturbation has amplified chemical contrasts in shallow versus deep waters, inducing primarily flushing patterns (concentration increase with streamflow) in agriculture lands and dilution (concentration decrease with streamflow) patterns in urban watersheds. Data and model reveal a general relationship between export patterns and shallow versus deep nitrate contrasts. This underscores the often-overlooked role of N distribution over depth. The results challenge commonly-held perception that legacy stores in agricultural lands induce chemostasis with negligible concentration variations with discharge. They suggest nutrient export will exacerbate as extreme events such as flooding intensify in the future climate.