|Has anyone ever had any experience using weather data? I've tried for over a month generating daily time-series of weather data. I am to create 2 files a parameter file (*.wg) which contains the parameters required to generate synthetic weather time-series and a statistics file(*.sta) which contains the seasonal frequency distributions. 1. I can find the Sitebase directory, but I am having a problem opening the data which is to automatically open up in |Notepad. I think the problem is that I'm not specifying the location and name of the file correctly? When it opens in Notepad it is supposed to be reading an Excel file of data? but all it is is gibberish? |
Probably too much right?
HERE'S PART OF THE ASSIGNMENT
Create future ‘synthetic weather’ using a weather generator downscaling tool
You will now learn how to use a statistical weather generator, LARS-WG, to create current and future synthetic weather time series based on the GCM scenario inputs that you obtained from the CCCSN.
1. Download and install LARS-WG; it can be accessed at http://www.rothamsted.ac.uk/mas-mod... If you haven’t already done so, follow the instructions to register your copy and apply for a personal license key.
2. Read the LARS-WG user manual (http://www.rothamsted.ac.uk/mas-models/download/LARS-WG-Manual.pdf). Note that the manual is for a previous version of the program, so some of the commands or interface may differ slightly; however it will help you understand the steps involved. The quick-start guide is also recommended: http://www.rothamsted.ac.uk/mas-mod...
3. After installation, open LARS-WG and go to ‘Options’. Ensure that the file paths for ‘Site Analysis’, ‘Generator’ and ‘Output’ correctly refer to the location in which LARS-WG was installed, while maintaining the ‘\Sitebase’ and ‘\Output’ subfolders in the paths. Click the red checkmark when you are finished.
The first step in generating weather data with LARS-WG is calibrating the model to your local climate station data with ‘Site Analysis’.
4. Follow the instructions in Section 3.1 of the user manual to prepare station information (.st) and weather (.dat) files for your station in notepad. Create the weather file in Excel, using the 1981-2010 daily climate records from your daily workbook in the previous half of this assignment. You will need min and max temperature as well as total precipitation (called ‘rain’ in LARS-WG).
Tips: Replace any blank/missing values with -99.0.
It is best that you remove the leap days (i.e., Feb 29) so that every year has 365 values. You will then need to create a new field called ‘JDAY’ that re-numbers the days in each year from 1 to 365, rather than starting at 1 with each new month. It is also critical that the days are sorted in chronological order starting with the earliest record.
When you are finished preparing your dataset, select all, copy, and paste into Notepad. Save as file type ‘All files’ and include .dat at the end of the filename.
5. Once you have prepared the two files, click on ‘Analysis’ and then ‘Site Analysis’ and select the .st file that you just created. Click the graph icon to run site analysis. If successful, LARS-WG will produce two new files for your station in the ‘Sitebase’ directory: a statistical characteristics (.stx) and weather generator parameters (.wgx) file.
>> Task #20: Prepare the required LARS-WG input files for your station and run Site Analysis.
After the model is calibrated, it can be validated using the ‘QTest’ described in Section 3.2 of the user manual. However, we will skip ahead to the ‘Generator’ step (refer to Section 3.3). The instructions for this part of LARS-WG differ slightly from those in the manual, but the basic steps remain the same.
6. Prepare a scenario (.sce) file based on the GCM experiment you selected in Task #18 and save it in the ‘Scenarios’ subfolder. Use the monthly change values that you calculated in Task #19. See the files “a1b-example-future.sce” and “a1b-example-nochange.sce” on LEARN for an example of recommended formatting.
Tips: Precipitation change should be in % (1.0 = no change, > 1 increase, < 1 decrease) and temperature change in degrees (0.0 = no change, + increase, – decrease).
Values for wet/dry spell, temperature variability and radiation should be left as 1 (no change), as we do not have these data.
Be sure to change the text beside ‘[NAME]’ in the .sce file to something unique and informative (e.g., identify the model, experiment and time period), as LARS-WG will append this to the end of the output filename.
7. Once you have prepared the .sce file, click on ‘Generator’ and then ‘Site’. Select your station from the dropdown list. Set the number of years to 50 and leave the random seed as is. Click ‘Baseline’ and then run the generator by clicking on the graph icon. This produces a 30-year daily time series of synthetic weather for the 1981-2010 baseline period (i.e., no climate change scenario is applied). You will find two files in the ‘Output’ folder named ‘stationnameWG.dat’ and ‘stationnameWG.st’ – the .dat file contains your time series data.
8. Repeat the previous step with the same criteria, but this time click ‘Scenario File’ and enter the path to your .sce file. Click the graph icon. You will now find two more files in the ‘Output’ folder named ‘stationname_scenarionameWG’. (If you didn’t add a unique identifier in your .sce file, then the output will instead use the default name and overwrite your baseline file).