Search Results

You are looking at 1 - 6 of 6 items for

  • Author or Editor: M. Todd Walter x
  • All content x
Clear All Modify Search
James O. Knighton, Arthur DeGaetano, and M. Todd Walter

Abstract

Watershed flooding is a function of meteorological and hydrologic catchment conditions. Climate change is anticipated to affect air temperature and precipitation patterns such as altered total precipitation, increased intensity, and shorter event durations in the northeastern United States. While significant work has been done to estimate future meteorological conditions, much is currently unknown about future changes to distributions of hydrologic state variables. High-resolution hydrologic simulations of Fall Creek (Tompkins County, New York), a small temperate watershed (324 km2) with seasonal snowmelt, are performed to evaluate future climate change impacts on flood hydrology. The effects of hydrologic state and environmental variables on river flood stage are isolated and the importance of groundwater elevation, unsaturated soil moisture, snowpack, and air temperature is demonstrated. It is shown that the temporal persistence of these hydrologic state variables allows for an influence on watershed flood hydrology for up to 20 days. Finally, six hypothetical climate change forcing scenarios are simulated to estimate the influence of catchment conditions on the watershed runoff response. The possibility of drier summers and wetter springs with a reduced winter snowpack in the Northeast is also simulated. These hydrologic changes influence flood discharge in the opposite direction as climate effects because of a reduced snowpack accumulation and melt time. Strong hydrologic state influence on flood discharge may be most attributable to increased air temperature and decreased precipitation. Hydrologic state variables may change both the location and shape of seasonal flood discharge distributions despite expected consistency in the shape of precipitation statistic distributions.

Full access
Rebecca D. Marjerison, M. Todd Walter, Patrick J. Sullivan, and Stephen J. Colucci

Abstract

Flash floods cause more fatalities than any other weather-related natural hazard and cause significant damage to property and infrastructure. It is important to understand the underlying processes that lead to these infrequent but high-consequence events. Accurately determining the locations of flash flood events can be difficult, which impedes comprehensive research of the phenomena. While some flash floods can be detected by automated means (e.g., streamflow gauges), flash floods (and other severe weather events) are generally based on human observations and may not reflect the actual distribution of event locations. The Storm Data–Storm Events Database, which is produced from National Weather Service reports, was used to locate reported flash floods within the forecast area of the Binghamton, New York, Weather Forecast Office between 2007 and 2013. The distribution of those reports was analyzed as a function of environmental variables associated with flood generation including slope, impervious area, soil saturated hydraulic conductivity k sat, representative rainfall intensity, and representative rainfall depth, as well as human population. A spatial conditional autoregressive model was used to test the hypothesis that flash flood reports are made more frequently in areas with higher populations, even when other flood-generating processes are considered. Slope, soil saturated hydraulic conductivity, and impervious area are significant predictors of flash flood reports. When population is added as a predictor, the model is similarly robust, but impervious area and k sat are no longer significant predictors. These results may challenge the assumption that flash flood reports are strongly biased by population.

Full access
M. Todd Walter, Paul Mutch, Christa D. Salmon, Donald K. McCool, and Lars O. Hedin

Abstract

Though chart recorder data are widely available for many environmental parameters over long periods of record, the nonorthogonal coordinate systems of many types of recorder charts complicates data extraction such that it is almost always manual; a tedious chore involving hand-counting tiny red boxes. Perhaps the most common data available on these types of charts are precipitation from weighing-type recording rain gauges. Unfortunately, because most digitizer software assumes the data being digitized have an orthogonal coordinate system, the nonorthogonality inherent in these charts prevents simple digitization of the data. If methodologies exist for electronically extracting rain gauge data, they are not widely known; recent texts still outline manual break-point tabulation in detail with no reference to any software or published methodologies for electronic data extraction. A simple procedure is presented here to transform digitized chart recorder data into meaningful, accurate data using readily available software. A comparison of manually read precipitation data and precipitation data extracted with a digitizer showed average differences of less than half the smallest marked increment on the charts for both precipitation amount and time. Differences were almost exclusively due to human bias in locating break points rather than true error.

Full access
James Knighton, Geoff Pleiss, Elizabeth Carter, Steven Lyon, M. Todd Walter, and Scott Steinschneider

Abstract

Current generation general circulation models (GCMs) simulate synoptic-scale climate state variables such as geopotential heights, specific humidity, and integrated vapor transport (IVT) more reliably than mesoscale precipitation. Statistical downscaling methods that condition precipitation on GCM-based, synoptic-scale climate features have shown promise in the reproduction of local precipitation. However, current approaches to climate-state-informed downscaling impose some limitations on the skill of precipitation reproduction, including hard clustering of climate modes into a discrete set of states, utilization of numerical clustering methodologies poorly suited to nonnormal data, and a tendency to focus on relationships to a limited set of large-scale climate modes. This study presents a methodology based on emerging machine learning techniques to develop global approximators of regional precipitation and discharge extremes given a suite of synoptic-scale climate state variables. Archetypal analysis is first used to define regional modes of winter and summer extreme precipitation and discharge across the eastern contiguous United States. A 2D convolution neural network (NN) is then used to predict the co-occurrence of the archetypes using 300- and 700-hPa geopotential heights, 300- and 700-hPa specific humidity, and IVT. Results suggest that 300-hPa geopotential height, 700-hPa specific humidity, and IVT yield the most reliable predictions, although with some important differences by season and region. Finally, we demonstrate that the trained activations of NN convolutional layers can be used to infer the causal pathways between synoptic-scale climate features and regional extremes.

Full access
John W. Recha, Johannes Lehmann, M. Todd Walter, Alice Pell, Louis Verchot, and Mark Johnson

Abstract

Tropical Africa is affected by intense land-use change, particularly forest conversion to agricultural land. In this study, the stream discharge of four small headwater catchments located within an area of 6 km2 in western Kenya was examined for 2 years (2007 and 2008). The four catchments cover a degradation gradient ranging from intact forest to agricultural land under maize cultivation for 5, 10, and 50 years. The runoff ratio (e.g., annual catchment discharge expressed as a percentage of rainfall) increased with increasing duration of cultivation from an average of 16.0% in the forest to 32.4% in the 50-yr-old agricultural catchment. Similarly, the average runoff ratio due to the stormflow component was 0.033 in the forest and increased gradually to 0.095 with increasing duration of cultivation. The conversion from forest to agricultural land in the first 5 years caused about half of the total observed increases in runoff ratio (46.3%) and discharge in relation to rainfall (50.6%). The other half of the changes in discharge occurred later during soil degradation after forest clearing. With increasing duration of cultivation, soil bulk density ρb at a depth of 0–0.1 m increased by 46%, while soil organic carbon (SOC) concentrations and total porosity decreased by 75% and 20%, respectively. The changes in hydrological responses that occurred in the initial years after forest clearing may suggest a significant potential for improved land management in alleviating runoff and enhanced storm flow and moisture retention in agricultural watersheds.

Full access
M. Todd Walter, Daniel S. Wilks, J-Yves Parlange, and Rebecca L. Schneider

Abstract

Recent research suggests that evapotranspiration (ET) rates have changed over the past 50 years; however, some studies conclude ET has increased, and others conclude that it has decreased. These studies were indirect, using long-term observations of air temperature, cloud cover, and pan evaporation as indices of potential and actual ET. This study considers the hydrological cycle more directly and uses published precipitation and stream discharge data for several large basins across the conterminous United States to show that ET rates have increased over the past 50 years. These results suggest that alternative explanations should be considered for environmental changes that previously have been interpreted in terms of decreasing large-scale ET rates.

Full access