|Home||Archived January 13, 2017||(i)|
The Data Grapher is a set of online tools that allow users to create customized graphs and tables of a whole variety of time-series data that are served up by the U.S. Geological Survey.
Now, if you've never been here before, it probably would be helpful to check out the Help system. Under the Help menu, there are a number of tutorials and example graphs. Under Example Graphs, you can kind of get a sense for the types of graphs that can be made with the Data Grapher system. These are time-series graphs of a couple different parameters, or comparing time series among different sites-- two or three sites, showing repeating patterns over years, making color maps, and so forth. So, that's a good place to start. The other thing you can do under the Help menu is to check out the video tutorials that have been set up. These tutorials show you how to use the navigation menu up here; they also tell you how to make up your first graph, and after you've made that graph, how you can make adjustments to it.
So why don't we start with the one-site Data Grapher, to visualize some time-series data. Now, the steps required in each of these various programs are very similar, and you start by choosing a river basin. In the Oregon Data Grapher, we have about 20 river basins set up because we're trying to manage about a thousand different time series from almost 300 different sites. It's easier to choose the site of interest if you first choose a river basin. So let's choose the Tualatin River in northwestern Oregon. If you do that, then you limit your search to only about a couple of dozen sites. We're going to choose the Tualatin River at Oswego Dam. Once you've chosen a site, it shows you the parameters that are available at that site, and the starting and ending dates for the data that are available. Next, you simply have to choose the type of graph, either a time-series graph or an X-Y plot. For this example, we'll choose a time series.
Then you choose the types of parameters you want to plot, so let's just choose dissolved oxygen and we'll choose oxygen percent saturation. Now, this is a good place to mention that you have the ability to choose a daily statistic. So, for example, if you're interested in the daily minimum dissolved oxygen concentration, you can choose that, and you can apply a running average. Now, in this particular case, for oxygen there's also this kind of odd option to not allow supersaturation. In this case, if we were to check all of these various things, we would end up with the 7-day running average of the daily minimum, with no credit for supersaturation, and that exactly corresponds to one of the water-quality standards in this particular river. The Data Grapher is nice because you can actually create statistics that correspond to water-quality standards. For now, though, let's turn these options off.
After you've chosen the parameters, you need to choose a date. Let's choose the entire year of, say, 2011. You can see that for this site, we have over twenty years of data set up and available for you to explore. But, let's just choose January through December of 2011, and then you say "Make Graph" and the results follow.
On the results screen, it kind of summarizes your inputs, gives you a couple of notes, and then shows you the graphical result. In this case, this is a time-series graph of dissolved oxygen in blue and oxygen percent saturation in red, on the right-hand scale, against an X axis of the date for an entire year. From the results screen, you can do a number of different things if you scroll down towards the bottom. You can either download the data used to create this graph. You can look at some inventory tables to show you the amount of data that is available. You can download the graph in a variety of different formats; so if you wanted to make a PDF, you can do that. You can also rescale the graph by entering different axis limits for the parameters that you're plotting. Or, you can rechoose the time limits for this site.
So, if we were to plot, for example, only the data for June of 2011, we would simply reselect those dates, check "Replot" and the result would be shown for just June. Here, we've zeroed in from a year all the way down to a month, and we can see very clearly the daily variations in some of these data. If we wanted this plot in a PDF format, say we were happy with it, we would just choose that format and click "Download" and up pops our PDF. It's pretty convenient.
One of the nice things that you can do with the Data Grapher is that you can compare data from among sites. So, let's return back to the main menu and let's choose a time-series option for two different sites. Using this, let's build a graph that compares water temperature at a couple of different sites. Just to keep things easy, let's stay in the Tualatin River basin. We'll keep the Oswego Diversion Dam as one of our sites, and for our second site, let's choose a site that's a little farther upstream. Once we have that, we'll just choose water temperaure for the two parameters, and let's go ahead and choose July and August of 2011 for our comparison. And, choose "Make Graph."
As you can see, the result pretty clearly shows a comparison of water temperature from two different sites in the same river basin, showing that the downstream site in blue is a little bit warmer during the summer because it has had a chance to warm up. This is a nice way to compare sites upstream and downstream of each other as well as sites in different river basins to see whether the patterns match each other, both for consistency of the comparison and just to see how the different rivers might behave a little differently.
Another way to use the Data Grapher that is pretty useful is to look at patterns over time, and one of the ways to do that is to choose the By Year/Month option. You can compare data from many years on the same axis. For this, let's choose a site in the McKenzie River basin that's downstream of a dam where they've changed the operations. Let's choose the South Fork McKenzie River near Rainbow, which is downstream of Cougar Dam. We'll just look at water temperature; in this case, let's choose the daily maximum with a 7-day running average because that's the water-quality standard that's used. Let's include most of the years-- so let's start in 2001 and go through 2011. Let's look at a full calendar year, and make a graph.
The results are going to show us that there's been a bit of a change in the pattern of the annual water temperature over the years. This line down here at the bottom shows the first year on the plot, 2001, before any operations at the dam were changed. In 2002 through 2004, the reservoir was drawn down behind the dam, and you can see that they had some temperatures that were a little warmer in September. But, then they changed operations in 2005 after a major construction project, and they've been able to create more of a natural seasonal temperature pattern since that time. This sort of graph allows you to show data with repeating patterns over a particular type of time period, which can be done either for the entire year or for specific months.
Another way to visualize repeating patterns in the data is to look at a color map. From the menu under Time Series, let's choose Color Maps, and let's choose a site in the Clackamas River basin. We'll take a site down near the mouth at Oregon City, and let's choose pH because I happen to know that that shows some nice patterns. Let's choose an entire calendar year in 2010. So, we just choose our dates-- make sure this goes from January through December. We want to plot the date on the X axis and the hour of day on the Y axis, and we have to choose a color scale. You have many different color scales to choose from, and I'm just going to choose this kind of standard blue-to-red color scale. It's very easy to reverse that, if you wish. Lower pHs on this color scale are going to be plotted with the blue and higher pHs in red. You can change the type of scale, but let's go with this for now.
This is the color map that results, and this is a nice way to visualize how the pH conditions have changed over the course of the year from January on the left to December on the right, and the hour of day from the top to the bottom. This shows you that the highest pHs always happen in the afternoon; they tend to happen in summer, and there have been times when the higher pHs have been interrupted. It turns out that this particular time period in the middle was a period of higher flow, as was this one over here, and so that when the algae are growing, to create higher pHs, they need a lower flow condition. The high flow disrupts the algal growth. This is a great way to visualize some of these patterns in the data.
Another way to kind of dig into the data and understand some of the processes that might be important is to take a look at specialized X-Y graphs. These are some of my favorite plots. Let's choose the Klamath River for this particular analysis, and the site-- let's choose the Miller Island site. We're going to plot water temperature on the X axis against dissolved oxygen on the Y axis. Because the solubility of dissolved oxygen is very much a function of water temperature, we have the option here of including oxygen percent saturation contours on that plot. That will be interesting, and you'll see that when it comes up at the end. Let's choose all of the data; it's just four years and so it doesn't take that long to plot. We're going to highlight some of the results by month. This allows us to plot different points in the graph with different colors depending on what month those points were measured.
Let's take a look at the result. You'll see that we get some pretty interesting patterns at this particular site. The graph that we get, again water temperature on the X axis and dissolved oxygen on the Y axis, with these contour lines showing constant percent saturation. The solubility of oxygen as a function of water temperature, if we neglect barometric pressure, is this thick dark line here. Anything above that line are oxygen concentrations that exceed the solubility; we call that "supersaturated," and the only way that that can usually happen is if there's enough photosynthesis going on in the river to produce an excess of oxygen. Similarly, values below that thick line are "subsaturated;" they are less than the solubility. Usually the way that that happens is that there are processes in the river that are consuming oxygen -- organic material that is being decomposed by bacteria can decrease the oxygen concentration. One of the cool things about this plot is that it shows an annual pattern. If you take a look at the different colors represented by the different months, you'll see that in the springtime, you're kind of moving towards the right along the solubility with a little bit of supersaturation showing that there's some algae activity, but you don't get a lot of algal activity until June or July, in the purple and kind of orange colors, but then at that point you also get a lot of oxygen demands, and the oxygen concentrations start going way down. The oxygen levels at this site actually will get very close to zero, and they don't recover in the fall and winter until very late after some of those oxygen demands have been depleted. This is a really cool way of looking to see at what time of year certain processes like algal growth or decomposition of organic material, along with solubility considerations, all come together, all of those influences into the dissolved oxygen budget of a stream. You can really understand how the system works a little better.
And of course, the Data Grapher and Data Tabler system allows you to actually access many of the data that are used to create all of these plots. Under Tables, you can actually create your own customized data tables. If you were interested in data from the Tualatin River at the Oswego Dam site, you just select the parameters of interest, select a daily statistic if you wish, or a running average. Choose a time period of interest, so we could just choose, say, the most recent month. Decide how you want to round those data. You have the choice of outputting data to the web browser, or you can send it to a file that might be easier to open up in a spreadsheet, and just make the table. It's all very simple, and there's the result.
So to sum things up, the Data Grapher is a set of online programs that allow anyone to explore and visualize USGS time-series datasets, make custom graphs and tables and compare datasets among sites and years. We think it's a great tool, and we hope you enjoy using it!
Many thanks to Steven Sobieszczyk for his expertise and assistance in creating this tutorial!
|Home||Archived January 13, 2017|