Frequently asked questions

Browse our frequently asked questions below. Need to know something else? Throw us an email or schedule a demo with our team today.

What is HydroForecast and how is it different?

HydroForecast is a software product that models inflow into dams and rivers. It is the only distributed deep-learning machine learning model in operation today.

  1. Methodology — learn more about our unique technology.
  2. Accuracy – jump over to our forecasts page to see the product in action.
  3. Usability – hear from our customers, explore the dashboard, and learn about our straightforward setup process.

What is a streamflow forecasting model?

Streamflow is a measure of flowing water, usually expressed as a rate, e.g. cubic feet per second. It tells us how much water will move through a single point (e.g., a given spot in a river). Sometimes streamflow is expressed as a volume over longer time horizons, e.g. total flow over a day or month, in acre-feet or cubic meters.

Streamflow forecasting models are built on an underlying hydrologic model that takes inputs from influential variables and accounts for complex interactions in the environment to output future flow predictions. These models can range from physically based to statistically based, and each has advantages. Standard hydrological model inputs include meteorological data, such as precipitation and temperature, land surface characteristics, e.g. elevation and slope.

What role does machine learning play in advancing the science of streamflow forecasting?

Excellent question! A specific type of machine learning model has recently been growing in usage within the hydrologic community, both in academic literature and in operational applications. Upstream Tech’s research has been a part of that growth, both with co-authored publications and in case studies that demonstrate customer use cases. This research paper laid the groundwork of benchmarking experiments showing the accuracy improvements using a machine learning prediction model over a traditional physically based model. 

Read our guide for those interested in learning more about machine learning in hydrology.

Briefly, a few advantages of machine learning models for streamflow prediction: 

  • Train a single model that can learn generalized hydrology, reducing the basin-by-basin approach and thus reducing the maintenance associated with model up-keep
  • Handle dynamic changes within a basin without recalibration; machine learning models can process data and build memories in ways that physically based models cannot
  • Experiment quickly with extremely fast run times and development cycles; distributed physically based models are computationally expensive and time consuming to run, making it slower for implementing new research

What is the relationship between climate change and streamflow?

This is a difficult question, and one that many scientists are trying to untangle and quantify. Generally, extreme storms, drought, and shifting rain patterns are testing the resiliency of our water infrastructure and challenging forecast models built for a more stable period. 

  • Historical records ≠ future conditions. Climate change means a future that is both wetter and drier, hotter and colder, with greater variability and less reliability.
  • Historical relationships between precipitation and runoff are shifting in ways that current models are having trouble capturing.
  • These realities point to the need for more flexible hydrologic models that can predict outside of the historical record from a single calibrated basin.

Read our post on why streamflow forecasts are so important.

What is theory-guided machine learning?

Unlike physical models, which use equations based on empirical studies of how interactions occur in the natural world, statistical models are a blank slate. They learn about relationships by combing through data and identifying signals. This can be problematic if not properly managed because coincidences can often be mistaken for patterns.

The idea behind theory-guided machine learning is to design a model that can take advantage of the benefits from both physical and statistical approaches. HydroForecast is our solution to this challenge for streamflow forecasts.

What time scale does HydroForecast operate on?

HydroForecast is designed to adapt to different needs. Different needs often call for a different time frame, or scope, of interest – what we call horizons. We offer 6 horizons that can be tailored fit to your organization. In brief, the 6 horizons:

  1. Historical: HydroForecast can construct a 20-year daily continuous streamflow record at any location in the world, partially gauged, or completely ungauged, and in both natural and altered watersheds.
  2. Virtual gauge: Increase the density of a streamflow monitoring network by adding points where having up to date weather, snow and streamflow data is important for real-time decisions.
  3. Short term: Hourly 10-day ahead forecast for operational decision-making – the most accurate in North America.
  4. Seasonal: Daily updating, 90-day ahead forecasts for proactive water and risk management.
  5. Extended seasonal: Need streamflow and volume information out to one-year ahead? This horizon extends the 90-day seasonal to cover the full 12 months ahead.
  6. Long term: Understand how water patterns could change for a specific area under different climate scenarios.