peterson-tim-j / hydrosight Goto Github PK

Please add the following feature to the TFN models:
Removal of fluctuations in groundwater level time series due to barometric pressure and Earth tides, and graphical presentation and verbal indication of cumulative response behaviour as per Rasmussen and Crawford 1997 Groundwater 35/3 pp. 502-511, Spane 2002 Water Resources Research 35/6 pp. 1-17, and Toll and Rasmussen 2007 Ground Water 45/1 pp. 101-105, to identify degree of aquifer confinement, lag time, other aquifer properties if possible, and well-bore storage and skin effects.
Request from Dr. Michael S. Smith, Hydrogeologist, NSW DPI-Water.

When exporting data prep. bores the export progress bar hangs. On inspection of the code, this appears to occur when bores have been deleted from the GUI data prep tab but remain within this.dataPrep list.

Peterson and Western, 2014, Nonlinear time-series modeling of head.pdf

The acronym GST is terrible. Possible alternate names are:
HydroSight
GeoData
...

It appears that the model requires forcing data to have starting date a few years earlier than the stating date of obshead. I tried forcing data with different starting dates. The one with 10 years worked, but the one with 5 years didn't. Not sure what the minimum is. Cheers, Xiang

I received the follow unexpected error:

Hi Tim,

I'm trying to use HydroSight Outlier Removal function to remove the outliers in my GW heads file.
However, the system prompts me that I've met a Data Analysis Algorighm failure.
I've attached the error screen shot, the model and the GW head file for your consideration.
Please don't hesitate to let me know if you need any further info.
Thank you so much for your help!

Cheers,
Xinyang

HydroSight.zip

Add a feature to the GUI to plot results specific to a model, for example the weighting functions for the TFN model or for the soil moisture model the moisture, runoff, drainage etc.

Shapoori_2015B.pdf

At example_TFN_model.m, line 72, the output flux option "drainage_normalised" seems to be an invalid output variable name from the forcing transformation function, which stops the algorithm at line 447 of model_TFN.m. Substituting "drainage_normalised" by "drainage" seems to solve it. Thanks

Please add a feature to display the calibration iterations inside the GUI Calibration RHS pane rather than the matlab command window.

Shapoori et al, 2015b,Estimating aquifer properties.pdf

Add a feature to plot the input forcing data for a built model. Ideally, the daily, monthly and annual data should be plotted.

Please add an option to view summary statistics of the calibrated head, observed head and input forcing. Such statistics should include the mean, median, 25 & 75%iles. For the daily forcing data this should also include the daily, monthly and annual and serial correlation.

when run the sentence "if run7paramModel ...... end", the calibration scheme is unknow.

Changing the width is difficult because it is difficult to locate the point between panels that changes the cursor, allowing the window widths to be changed.

I found it is important to have >=2 periods -999 values ending with metered volume (rather than 1 in the Wiki document) when pumping input data is prepared. If you put in all periods of -999 values ending with no metering (0 pumping rate) or one of them ending with a metered volume, the model will produce negative downscaled pumping rates. Why?

Hi Tim,

I met issue in the simulation with HydroSight.
I created several simulation in one project. However, only the FIRST simulation can be shown on the right-hand side table and be exported. When I click on my OTHER simulations, the software doesn't react.. That is, I cannot see and export my OTHER simulations.
Could you please kindly have a check? I've attached my model, input files and screen shot for your consideration. Thank you so much for your help!

Cheers,
Xinyang

SensitivityAnalysis.zip

Please consider including the following functions:

1. Cumulative monthly residual rainfall on same graph as groundwater level time series.
2. Statistics to compare the likeness and lag of these two plots.
   Request from Dr. Michael S. Smith, Hydrogeologist, NSW DPI-Water.

The derived value showing the lag time to the peak weighting is incorrect.

Hi Tim,
Congratulation on the release of the HydroSight version 1.30.
This is just a question. I am trying to understand the concept of the downscaling and in-filling of infrequently metered pumping in the HydroSight. Have you put the documentation online?
To my understanding: Firstly the model estimates the average daily pumping rates for the pumping seasons with metered volume using total metered volume, the number of defined pumping days and groundwater hydrograph. Secondly the daily average of all metered seasons is applied to non-metered seasons. For those non-metered seasons, the daily pumping rates are the same. Pump on or off is determined by model calibration. Am I wrong?

Add icons to the GUI. These should include general icons such as open, save, save as plus plotting icons when a plot is displayed.

"Model build failed - Abstract classes cannot be instantiated. Class 'derivedForcing_linearUnconstrainedScaling_dup' inherits abstract methods or properties but does not implement them. Use meta.abstractDetails to see the list of methods and properties that 'derivedForcing_linearUnconstrainedScaling_dup' must implement if you do not intend the class to be abstract." occurred when I try to built the example of TFN model - landuse change with model label of "Reveg Catchment - Recharge and ET landuse impact" in Graphical User Interface.
Thank you!

HydroSight GUI does not open using matlab 2018b+. This is caused by changes to the plotting toolbar.

Dear Tim,

just to let you know that I had issues trying to calibrate models that were using GW observation data that was not listed chronologically. After correcting the input data, the issue stopped.

Having duplicate site ID names within the coordinates file causes unhandled errors when building a model. Duplicate IDs should not be allowed.

Hi Tim,

I've tried to use the newest version of HydroSight to make groundwater hydrographs.
The calibration has been run successfully. However, the result cannot be displayed in the GUI.
I've attached the error message and my model for your consideration.
Thank you so much for your help!

Best,
Xinyang

SensitivityAnalysis.zip

Projects with 1,000+ models is slow to load. This is primarily the issue when the models are off loaded to the HDD.

Shapoori_2015A.pdf

climate_only_tutorial.zip

When the TFN pearson's function parameter 'n' is less than 1 (when back transformed), the integration from t = 0 to 1 is problematic. That is, the theta function asymptotes to the y=axis and the analytical integration results in the first fraction of the first day to have an exceedingly large contribution.

Move existing HTML help from inside the GUI to the GitHub project page. This will allow more efficient navigation of the help documents.

Please consider including the following functions:

1. Cumulative monthly residual rainfall on same graph as groundwater level time series.
2. Statistics to compare the likeness and lag of these two plots.
3. Removal of fluctuations in groundwater level time series due to barometric pressure and Earth tides, and graphical presentation and verbal indication of cumulative response behaviour as per Rasmussen and Crawford 1997 Groundwater 35/3 pp. 502-511, Spane 2002 Water Resources Research 35/6 pp. 1-17, and Toll and Rasmussen 2007 Ground Water 45/1 pp. 101-105, to identify degree of aquifer confinement, lag time, other aquifer properties if possible, and well-bore storage and skin effects.
   Dr. Michael S. Smith, Hydrogeologist, NSW DPI-Water.

A forcing data file where each row ends with a comma causes a blank column to be input, with the outcome that the model cannot be built. This should be fixed.