Search for: All records

Nitrogen (N) is a critical element in many ecological and biogeochemical processes in forest ecosystems. Cycling of N is sensitive to changes in climate, atmospheric carbon dioxide (CO2) concentrations, and air pollution. Streamwater nitrate draining a forested ecosystem can indicate how an ecosystem is responding to these changes. We observed a pulse in streamwater nitrate concentration and export at a long-term forest research site in eastern North America that resulted in a 10-fold increase in nitrate export compared to observations over the prior decade. The pulse in streamwater nitrate occurred in a reference catchment in the 2013 water year, but was not associated with a distinct disturbance event. We analyzed a suite of environmental variables to explore possible causes. The correlation between each environmental variable and streamwater nitrate concentration was consistently higher when we accounted for the antecedent conditions of the variable prior to a given streamwater observation. In most cases, the optimal antecedent period exceeded two years. We assessed the most important variables for predicting streamwater nitrate concentration by training a machine learning model to predict streamwater nitrate concentration in the years preceding and during the streamwater nitrate pulse. The results of the correlation and machine learning analyses suggest that the pulsed increase in streamwater nitrate resulted from both (1) decreased plant uptake due to lower terrestrial gross primary production, possibly due to increased soil frost or reduced solar radiation or both; and (2) increased net N mineralization and nitrification due to warm temperatures from 2010 to 2013. Additionally, variables associated with hydrological transport of nitrate, such as maximum stream discharge, emerged as important, suggesting that hydrology played a role in the pulse. Overall, our analyses indicate that the streamwater nitrate pulse was caused by a combination of factors that occurred in the years prior to the pulse, not a single disturbance event. 

This dataset consists of chemical analyses of subsurface water samples collected from Watershed 3, Hubbard Brook Experimental Forest, Woodstock, NH, USA from 2009-2020. Samples include groundwater samples pumped from monitoring wells, grab samples of natural groundwater seeps, and soil water samples pumped from Prenart lysimeters. For samples from wells where water table was monitored, depth to water table is given. These data were gathered as part of the Hubbard Brook Ecosystem Study (HBES). The HBES is a collaborative effort at the Hubbard Brook Experimental Forest, which is operated and maintained by the USDA Forest Service, Northern Research Station. 

Overstory foliage is collected in late summer from a reference forest to the west of Watershed 6 (also referred to as Bear Brook Watershed). Concentrations of C, N, P, K, Ca, Mg, and the natural abundance of N and C isotopes (delta-15N and delta-13C) in foliage are measured. These measurements, in combination with litterfall estimates of foliar biomass, allow us to estimate the pool of nutrients in foliage. They also allow us to estimate nutrient retranslocation, using measurements of leaf litterfall chemistry. Long-term measurements continue with the aim of detecting disturbances in nutrient cycling and trends in foliar chemistry over long time scales. These data were gathered as part of the Hubbard Brook Ecosystem Study (HBES). The HBES is a collaborative effort at the Hubbard Brook Experimental Forest, which is operated and maintained by the USDA Forest Service, Northern Research Station. 

This dataset consists of chemical analyses of subsurface water samples collected from Watershed 3, Hubbard Brook Experimental Forest, Woodstock, NH, USA from 2009-2015. Samples include groundwater samples pumped from monitoring wells, grab samples of natural groundwater seeps, and soil water samples pumped from Prenart lysimeters. For samples from wells where water table was monitored, depth to water table is given. These data were gathered as part of the Hubbard Brook Ecosystem Study (HBES). The HBES is a collaborative effort at the Hubbard Brook Experimental Forest, which is operated and maintained by the USDA Forest Service, Northern Research Station. 

This dataset consists of chemical analyses of subsurface water samples collected from Watershed 3, Hubbard Brook Experimental Forest, Woodstock, NH, USA from 2009-2015. Samples include groundwater samples pumped from monitoring wells, grab samples of natural groundwater seeps, and soil water samples pumped from Prenart lysimeters. For samples from wells where water table was monitored, depth to water table is given. These data were gathered as part of the Hubbard Brook Ecosystem Study (HBES). The HBES is a collaborative effort at the Hubbard Brook Experimental Forest, which is operated and maintained by the USDA Forest Service, Northern Research Station. 

Stream solute monitoring has produced many insights into ecosystem and Earth system functions. Although new sensors have provided novel information about the fine‐scale temporal variation of some stream water solutes, we lack adequate sensor technology to gain the same insights for many other solutes. We used two machine learning algorithms – Support Vector Machine and Random Forest – to predict concentrations at 15‐min resolution for 10 solutes, of which eight lack specific sensors. The algorithms were trained with data from intensive stream sensing and manual stream sampling (weekly) for four full years in a hydrologic reference stream within the Hubbard Brook Experimental Forest in New Hampshire, USA. The Random Forest algorithm was slightly better at predicting solute concentrations than the Support Vector Machine algorithm (Nash‐Sutcliffe efficiencies ranged from 0.35 to 0.78 for Random Forest compared to 0.29 to 0.79 for Support Vector Machine). Solute predictions were most sensitive to the removal of fluorescent dissolved organic matter, pH and specific conductance as independent variables for both algorithms, and least sensitive to dissolved oxygen and turbidity. The predicted concentrations of calcium and monomeric aluminium were used to estimate catchment solute yield, which changed most dramatically for aluminium because it concentrates with stream discharge. These results show great promise for using a combined approach of stream sensing and intensive stream discrete sampling to build information about the high‐frequency variation of solutes for which an appropriate sensor or proxy is not available.