Snowbedo

Package for modeling snowmelt and streamflow in semi-arid, snowpack-driven mountainous watersheds.

This package utilizes streamflow, temperature data, precipitation, shortwave incoming (downwelling) radiation, and white-sky albedo to calibrate a streamflow model based on a simplified snowmelt-based process that is influenced by albedo. This calibrated model can then be used to explore various scenarios of albedo (which may be influenced by dust and black carbon deposition).

Functions included in the Package

Formatting data

• limitperiod (limits data to specific time period)
• dissipate (calculates daily precipitation from accumulative precipitation)

Modeling streamflow

• modelflow (calculates flow based on meteorological and remotely sensed land cover parameters). Note that this function is still under development, and has not produced acceptable streamflow results.

Scripts for Using the Snowbedo Package

• NeuralNetwork.R (reads in the formatted data for an individual watershed, subsets based on user-defined list of parameters, and trains a neural network model for streamflow. Produces a time series of modeled streamflow and saves a file of the neural network model)
• ModelDifferencesInAlbedo.R (reads in a formatted dataset and neural network model (produced by NeuralNetwork.R) and creates different scenarios by adjusting the albedo. Produces time series of streamflow for each of the scenarios)