6 and 3 76 h ( Table 2) Modeled flow compared reasonably well to

6 and 3.76 h ( Table 2). Modeled flow compared reasonably well to observed flow at nine of the 10 gauges (Table 3 and Fig. 7). These values

improved during time periods when there was a rain gauge inside the watershed. For example, the Wappinger Creek NSE improved to 0.64 from 0.57 for daily flow after 2004, when a NOAA gauge is active inside the watershed. Using these measures, the model appears acceptable in nine of the 10 watersheds, although it 3-MA cell line fails in the Neshanic River, NJ in all three metrics. Interestingly, event flow analyses showed better performance relative to daily for the small watersheds and no change or worse performance for the larger watersheds. Over the 6-month period of observations in Town Brook watershed, the model predicted 16 occurrences of overland runoff. During 15 of those events, the median water table depth for locations estimated as being “wet” was less than 100 mm from the soil surface, while the median dry wells remained at or below a depth of 100 mm during all events (Fig. 8). This corroborates previous findings that overland runoff in the Northeast is initiated once the water table is within approximately 100 mm of the surface (Lyon et al., 2006 and Dahlke et al., 2012). Over the course of the 16 events, we compared 288 separate predictions

of wet or dry conditions to field measurements. In 18 cases (6%), we predicted a well to be wet when the water table at that location was below 100 mm and in 55 (19%) cases we predicted a well to be dry when the water table depth was within 100 mm of the PR 171 soil surface. The remaining 215 (75%) predictions correctly identified a location as wet or dry based on modeled results. On days when no runoff was predicted,

the average depth to the water table of all wells was 240 mm. At the Fall Creek site, four out of the 13 measurement dates were predicted to have Resminostat saturated areas contributing to storm runoff. In three of the four dates, the median volumetric soil moisture reading in the modeled wet locations was above saturation (i.e., ≥53%), while dry locations had median values below saturation (Fig. 9, top). On the date that the “wet” wells were below the saturated value (June 26, 2013), the observed streamflow at the outlet did not show a discernible rise in the hydrograph, highlighting the difficulty in correctly modeling small storm runoff events. The Cascadilla Creek site only had one instance of measurements being taken in locations predicted to be wet, and on this date, the wet sites had a median soil moisture status above saturation (Fig. 9, bottom). The model presented here shows promise as a simple tool allowing for spatial prediction of saturation-excess runoff locations in the northeastern US. Areas that were predicted to generate overland runoff had higher average soil moisture status and an elevated water table compared to areas modeled to be dry within three watersheds.

This entry was posted in Antibody. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>