Freshwater ponds are prevalent globally and provide critical ecosystem functions (e.g., water storage and groundwater recharge), yet little is known of their hydrological features immediately following formation.  We analysed stable isotopes of water (δ²H, δ¹⁸O) to characterize spatio-temporal hydrologic variation in ponds created by the Mount St. Helens eruption.  We also examined how climate and landscape features interact to regulate local hydrology.  Ponds were sampled for isotopic analysis in spring (2015, 2017, 2018) and summer (2018).  Mass balance models characterized the water balance of ponds (evaporation: inflow; E:I), as well as water sources (rain, snow).  Other variables were measured in situ (temperature, conductance), collected from data sources (meteorology), or quantified with remote sensing (vegetation).  Bayesian estimates of standard ellipse areas (SEA_B) were used to compare isotopic values among years, whereas linear models were used to examine local and regional drivers of E:I, as well as intra-annual isotopic shifts.  We observed high inter-annual variability (as SEA_B), suggesting that snow was the main water source in wet years, but that the proportion of rain and snow varied among sites in dry years.  Spring E:I was negatively correlated with total precipitation, whereas the importance of evaporation in summer varied with pond morphology, with large shallow ponds exhibiting the greatest evaporation. Evaporation regulated the hydrology of ponds with higher dissolved organic carbon (DOC; as residuals of DOC and chlorophyll).  We show that simple metrics of basin morphometry can predict seasonal variability in pond hydrology, allowing managers to better estimate pond sensitivity to future climate conditions.

There are two datasets included: 1) environmental data; and 2) isotope data. 

Variables in the environmental dataset were collected several ways, including in situ measurements (water temperature, specific conductance), analyzed in a laboratory following sampling (dissolved organic carbon), compiled from outside data sources (meteorological variables), or quantified with remote sensing (normalized difference vegetation index).  Data processing is described in the manuscript.

Variables in the isotope dataset were analyzed in a laboratory following sampling (δ²H and δ¹⁸O), compiled from outside data sources (meteorological variables), or represent intermediate calculations.  Briefly, water samples were collected by immersing high-density polyethylene bottles, which were then sealed tightly to prevent water evaporation.  Samples were filtered through glass fiber filters (VWR, 0.45-µm pore size) on the day of collection, and the filtrate frozen until for further chemical analysis.  Water samples were analyzed for δ²H and δ¹⁸O values using a Picarro L2120-I cavity ring-down spectrometer (CRDS), at the Institute of Environmental Change and Society (IECS), University of Regina, Saskatchewan, Canada (http://www.iecs-uregina.ca/).  All isotope results are reported in δ notation in per mil units (‰) relative to local and international standards, including Vienna Standard Mean Ocean Water 2 (VSMOW2) and Standard Light Antarctic Precipitation 2 (SLAP2).