Program to Simulate Rain+Ground Water

  1. Jul 22, 2014 #1
    The reason:
    I live in a small waterside town and we are very concerned over our water table. The problem is that we don't have a lot,if any, research or organized data to back up important decisions. The problem is that no one is informed on what they're talking about, many regulations are based on hypotheses, opinions, and bias. I'm interested in making a model for some of the hot topics:
    • Catching rain water
    • Septic contamination
    • Fertilizer/pesticide contamination
    • Filter methods
    • (Costal erosion if possible)

    What I need:
    I need a program that can simulate fluid dynamics with at least the following variables:
    • time
    • flow rate/porosity
    • rate of decay
    • can simulate rain (intensity + duration)
    • rate diffusion
    • color of fluids(to track them)

    • 3d visual/graphic display
    • ability to over lay the following maps and account for variables:
      • topographic (contours->water flow)
      • water table topography (depth and fluctuations)
      • sediment map (porosity)
      • microbial map(depth+decay rate of contaminate)
    • simulate wells(rate of water out+depth+location)
    • septic systems(rate of water and contaminates in+depth+location)
    • penetration and filtration of salt water
    • shows me water quality at test points or where I click

    If I haven't accounted for anything important please let me know!
    These are just the ideal needs, if you know of anything similar, I would really appreciate it!

    I understand it won't be perfect, but it should give an idea of where to look for problems. hopefully spectrometry, gel electrophoresis, and chemical testing can provide enough evidence to backup or revise our regulations.
  2. jcsd
  3. Jul 28, 2014 #2
    I suppose using Mathematica will help you make brilliant simulations if you get the dynamics and modeling right. Though it will need some awesome coding skills.
  4. Jul 30, 2014 #3
    Do u think a game engine might work?
  5. Jul 30, 2014 #4
    Yes! maybe, "Unity" might be helpful!
  6. Jul 30, 2014 #5
    I wish I had more time (and...frankly, inclination) to respond to this in the depth I feel I should, but I don't. So here's a succinct version.

    0. Everything I say here should be taken with a grain of salt. I'm an internet stranger, and though I intend for what I say to be helpful to you, I don't know you or your situation, so maybe my advice isn't the best for you in your specific situation. Also, don't take what I say to be discouraging. If you have a real drive to get something going, don't let my words stop you. I'm writing only to offer advice.

    1. You should know, right off the bat, that each of those five areas of simulation you are looking for are pretty complex (some extremely, some a bit less). Studies for each of these, individually, can take dozens, hundreds, even thousands of manhours to complete; and that is with existing, proven software.

    The complexity increases as you make it more realistic and empiric. Each of these will require a number of assumptions; how many assumptions you make and in what areas will affect, and be determined by, how you want to optimize/balance the accuracy desired with the time and money required to obtain accurate data.

    For instance. "Fertilizer contamination". How much fertilizer is being used? Do you make an assumption based on average usage per farm/household and assume usage over some percentage of the arable land? Is it constant? Seasonal? How permeable is the soil? Are there compounds that may collect the pesticides/fertilizer? Are there compounds that break it down? What percentage of the chemicals actually make it into the groundwater in the first place? These are just a tiny fraction of the variables you have to consider for just this one option to create a model with any usable degree of accuracy.

    2. Data. How will you collect data? Do you have wells that can be sampled? A lot of them? You'll need many wells to determine the porosity/permeability of the water, the water table level (based on topography), etc. You'll also need to test the water for various compounds. Then, how often will you test? Not only does water quality vary year by year, but also by season, and even by day in many cases. Especially if this is a rural area where water usage and fertilizer are seasonal.

    3. How will you collect data on existing systems like sewer and storm water drainage conditions, septic system conditions, well condtions, etc? How about data on contaminants from other areas? If your water table extends to other towns/areas/regions, how will you differentiate external contaminants from internal?

    4. Capabilities:

    This is where it gets tricky. Software which currently handles some of these items take years to develop into properly functioning interfaces. Of course, if you have a platform and work off that, and don't have to worry about user interfaces and frivolous functions, then you can save lots of time. Still, incorporating multiple types of software simulations together is tricky business. 3D integration is usually pretty straightforward, your simulations will have parametric elements to them anyway, so that's mostly visual. But incorporating test points, and multiple inputs and outputs may prove somewhat more difficult.

    I think it could be cool, and certainly useful. But really, the majority of these items are simply detailed in reports, tables, and drawings. See this webpage for a brief, really high level look at ground water reports .

    Here's a link to a page I found with some notes on existing aquifer modeling tools:

    The main issue you face is cost. It costs money to sample wells, to run tests, to interpret data, etc. If you're going to use this for your city's water, then you should be working with experienced hydrogeologists and people who work with groundwater pollution.

    A game engine might be helpful for the physics, and a powerful machine to run it might be able to get you what you're looking for, but none of that is worth anything unless you've got a good model mathematically. For that you must understand your soil conditions very well over a large area, and must know exactly what assumptions are being made and how that will affect your results. Otherwise, all you have is a pretty model...
    Good luck with however you decided to go about this. I think it'd be pretty cool, but practically speaking, a few charts and a couple section-view charts showing aquifer levels and contamination plumes would be cheaper and more conventional.

    Honestly, a thorough study of the existing system laid out in a report and coupled with good data plotted/drawn for meaningful metrics would be far more useful. This is something that could be done in a reasonable time and with good enough accuracy and utility that real changes could be made, positively affecting the community.
  7. Oct 20, 2014 #6
    you've got a good model mathematically. This is something that could be done in a reasonable time and with good enough accuracy and utility that real changes could be made, positively affecting the community.
  8. Oct 20, 2014 #7
    The results will certainly be dated, but if your area is in the United States, then check into the reports and studies prepared during the "208" program in the middle 1970's. The proper name of this program was the "Areawide Water Quality Management Program". Not all of these programs produced groundwater studies, but some did. As I said, the studies will be dated, but they will make an excellent point of departure.

    I directed such a program for a three-county area on the Lake Michigan Shoreline in Michigan. At that time, the MSGC analyses for wells ran about $3,000.00 per sample. It is probably more today.

    What you want would certainly be useful. However, I believe that it would be well beyond the financial means of most small communities. Back in the 1970's the EPA spent over four-million dollars and thirty man-years of time just for our three counties.
  9. Oct 22, 2014 #8


    User Avatar
    Science Advisor

    You don't want much do you! I guess you will probably need to write your own code using a finite element method. I wrote a simple one 40 years ago so it will be much easier now.

    The first thing to do is to put a number of sampling tubes in bores that can be used to sample ground water and measure the water table height. Flow rates of streams and wells should also be recorded. The sooner you start gathering that real data, the better can be your later modelling based on a longer baseline and trends. Early ground truth is far more important than finding or quickly writing the modelling program.

    If your local topography is flat you may find salt water coming inland as rainwater is captured or fresh water is extracted from wells. Salt will show up as high conductivity readings. If the land is steep the local geology and fracture patterns will be critical to water flow. The soil profile and vegetation will have a significant effect on the behaviour of rain versus evaporation.

    We have no idea of your rainfall, landuse, soil, geology or topography, so it is difficult to identify the complexity that will be needed in the software to produce verifiable predictions.
  10. Oct 23, 2014 #9
    Hi there, i'm a hydrogeologist and what your talking about has been done many times, and the basic results from those other studies can be applied to get to your end goal, as opposed to doing complicated modeling and time intensive sampling and costly analytical work etc. Typically something like this would be handled with a groundwater flow and contaminant transport model (GW vistas, MOD Flow). Operating, calibrating and interpreting groundwater flow modeling is complex and no something that you can really "just pickup" on your own without extensive knowledge.

    Also there are some way to get good reliable data without using analytical labs, i.e. field test kits for nitrate (fertilizer and septic system contaminant), conductivity meters (salt water intrusions) - all things that any hydrogeologist would be familiar with.

    I would recommend getting in touch with your State's geological survey, they always have a hydrogeologist or water specialist, or know who to direct you too. I would suspect that they can provide some very useful help. Of course you could call a hydrogeology/engineering consulting company, which would result in major salivation on their part and a lengthy proposal being delivered to you... but your talking huge bucks for a detail analysis like your talking. The would suggest that a better strategy is to use the State resources, common sense (provided by geologists or engineers), and apply the results from other studies / town ordinances to your situation.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted
Similar Discussions: Program to Simulate Rain+Ground Water
  1. Acid rain (Replies: 8)

  2. Germs cause rain (Replies: 1)

  3. Can we stop rain? (Replies: 12)