Estimating Drought Streamflow Probabilities for Virginia Streams
Planning for drought conditions in Virginia streams is essential to the sound management of water resources and associated riparian and watershed ecosystems. Reliable estimations of the likelihood that stream flows during drought-prone months will exceed specific low-flow thresholds can provide advance warning of drought conditions, allowing extended lead times for improved drought awareness and effective management response.
Improved knowledge and estimation of low flows in drought-prone months provide extended lead time for drought response even as precipitation, water withdrawals, land uses, and climate variables change over time. Drought stream flow probability estimates can be used for the management of riparian systems and the ecology associated with forested, agricultural, and urban landscapes. Drought stream flow probability estimates provide a basis for analysis of future streamflow response to changes in ecosystem and climate.
Probabilities of streamflows exceeding specific drought streamflow thresholds are characterized in a set of 70,120 maximum likelihood logistic regression (MLLR) equations for 339 basins in Virginia. A 46,704-member subset of these equations is published for 259 Virginia basins. More than 5.12 million streamflow daily values, collected over the period of record, were compiled and used in the analysis. Example calculations are available demonstrating how to use the equations to estimate probable streamflows up to 8 months in advance.
Results
MLLR equations are provided that predict the chance of exceeding or not exceeding one of a set of drought streamflow thresholds during the historically dry months of the year—July, August, and September—each a function of mean streamflow values from the previous historically wet months of the year—November, December, January, and February. Because equations are provided for each of two probability categories associated with each response variable, calculations are easy to make and easy to interpret. For a given streamgaging station, an equation may be chosen that addresses the probability of exceeding a streamflow threshold value of interest P[Yes] or not exceeding a streamflow threshold value of interest P[No] as explained by mean monthly streamflow from one of the preceding winter months (November, December, January, or February) or the averaged mean monthly streamflow of these four months.
Publications describing drought streamflow in Virginia.
Methods for estimating drought streamflow probabilities for Virginia streams
Low-flow characteristics of Virginia streams
Below are partners associated with this project.
Planning for drought conditions in Virginia streams is essential to the sound management of water resources and associated riparian and watershed ecosystems. Reliable estimations of the likelihood that stream flows during drought-prone months will exceed specific low-flow thresholds can provide advance warning of drought conditions, allowing extended lead times for improved drought awareness and effective management response.
Improved knowledge and estimation of low flows in drought-prone months provide extended lead time for drought response even as precipitation, water withdrawals, land uses, and climate variables change over time. Drought stream flow probability estimates can be used for the management of riparian systems and the ecology associated with forested, agricultural, and urban landscapes. Drought stream flow probability estimates provide a basis for analysis of future streamflow response to changes in ecosystem and climate.
Probabilities of streamflows exceeding specific drought streamflow thresholds are characterized in a set of 70,120 maximum likelihood logistic regression (MLLR) equations for 339 basins in Virginia. A 46,704-member subset of these equations is published for 259 Virginia basins. More than 5.12 million streamflow daily values, collected over the period of record, were compiled and used in the analysis. Example calculations are available demonstrating how to use the equations to estimate probable streamflows up to 8 months in advance.
Results
MLLR equations are provided that predict the chance of exceeding or not exceeding one of a set of drought streamflow thresholds during the historically dry months of the year—July, August, and September—each a function of mean streamflow values from the previous historically wet months of the year—November, December, January, and February. Because equations are provided for each of two probability categories associated with each response variable, calculations are easy to make and easy to interpret. For a given streamgaging station, an equation may be chosen that addresses the probability of exceeding a streamflow threshold value of interest P[Yes] or not exceeding a streamflow threshold value of interest P[No] as explained by mean monthly streamflow from one of the preceding winter months (November, December, January, or February) or the averaged mean monthly streamflow of these four months.
Publications describing drought streamflow in Virginia.
Methods for estimating drought streamflow probabilities for Virginia streams
Low-flow characteristics of Virginia streams
Below are partners associated with this project.