Subject: stream flow projection under climate change scenarios Posted: 1/9/2017 Viewed: 4551 times
I have two questions about simulation of future hydrological variables by the use of WEAP.
1. I have the climatic data of a station and I used them to project the future temperature and precipitation based on the fifth assessment report of IPCC scenarios. Now I have current climatic data(temperature and precipitation) and also current stream flow data for the period of 1980-2010. On the other I projected future precipitation and temperature(2011-2040) by the use of SDSM downscaling model. Now I want to know that how I can simulate future stream flow by the use of these data? I studies the "hydrology" part of the user guide but unfortunately I didn't get it very well.
2. Can I use these data to simulate future changes in underground storage? I mean is WEAP able to predict future changes of groundwater resources in associated with precipitation and temperature future changes?
Thank you for your considration
Subject: Re: stream flow projection under climate change scenarios Posted: 1/10/2017 Viewed: 4516 times
If you have climate data, you can generate streamflow using WEAP catchments (for temperature and precipitation data, I suggest using the soil moisture catchment methodology that is detailed in the Hydrology chapter of the tutorial). Note that temperature and precipitation are two important climatic components of this, but there are others (humidity, windspeed, cloudiness fraction). WEAP has default values for these, if you don't have data, and that's a fine place to start.
The key is that the soil moisture model also has land use parameters that need to be quantified but are typically very difficult to measure. So we usually determine them through calibration. This means:
1) Set your model for the historic period
2) Plug in your historic climate data and your historic streamflow gauge
3) Change the land use parameters in the catchment (runoff resistance factor, soil water capacity, etc) so that the following is true:
a) The modeled and observed streamflow are as close as possible (we recommend using the Nash Sutcliffe statistic to quantify this)
b) The catchment Relative Soil Moisture 1% shows the same seasonal pattern each year (not growing or shrinking)
c) the catchment Relative Soil Moisture 2% is steady over time, possibly with some small seasonal pattern
You can review the tutorial chapter about Data and Formatting to see how to compare observed and modeled streamflow in the Results. We are hoping to produce a tutorial chapter about the calibration process this year.
Once you have calibrated your historic model, you can switch the years of the model to the future, with the future climate plugged in, and let the same hydrological processes that you calibrated to represent the past now represent the future based on the climate inputs.
You can simulate future groundwater changes, again using the Soil Moisture methodology for the catchment. This models 2 "buckets" - a top bucket of soil moisture, and "lower bucket" that is basically the groundwater part of the hydrologic system. You can model this either within the catchment, or connect the catchment to a groundwater node and then the groundwater node becomes the "lower bucket." The Hydrology tutorial shows how to do this.
But remember, any model that claims to have accurate (and therefore useful for planning purposes) predictions of the future must first demonstrate accurate representations of the past though calibration and reporting of the statistics such as the Nash-Sutcliffe value.
Topic "stream flow projection under climate change scenarios"