Water Movement in Turfgrass Soils

Luqi Li Author

08/06/2020 Added

25 Plays

Description

Dallas Williams - Water Movement in Turfgrass Soils

Searchable Transcript

Search:

[00:00:00.623](upbeat music)
[00:00:12.670]Hello everyone.
[00:00:14.050]Thank you for joining me and taking the time
[00:00:16.160]to listen to my field day presentation.
[00:00:18.610]I'm Dallas Williams, a grad student.
[00:00:21.030]I'm in the Turfgrass Department here at UNL
[00:00:23.620]under the advisement of Roch Gaussoin.
[00:00:26.840]My research has to do with finding a method
[00:00:29.690]to trace and calculate water movement in turfgrass soils,
[00:00:33.310]more specifically golf course putting greens
[00:00:35.860]is the type of profile we've been working with.
[00:00:39.850]For those of you who attended field day last year,
[00:00:42.390]you may remember me talking about
[00:00:44.080]some of the challenges we were facing with the project.
[00:00:47.260]We were having a lot of different soil probes being built,
[00:00:50.110]we were having problems
[00:00:51.310]with uniform representative infiltration.
[00:00:55.625]In this presentation I will share with you
[00:00:58.230]all that we've been working on the past year
[00:01:00.130]with our algorithm development and software work.
[00:01:06.300]So I'll first start by asking,
[00:01:09.100]why do we care about water movement in turfgrass soils?
[00:01:14.100]Well, proper water infiltration and percolation
[00:01:16.700]is important for several different things.
[00:01:19.830]Fertilizer and irrigation efficacy certainly,
[00:01:23.750]a lot of fertilizer and water can be lost due to run off
[00:01:27.270]if it's not being infiltrated
[00:01:29.350]and accepted into the soil profile.
[00:01:32.520]Plant available water is certainly determined
[00:01:34.990]by infiltration and percolation
[00:01:37.630]and this in turn relates to turf health.
[00:01:41.982]Our management practices,
[00:01:43.720]as well as playability and cultural use are also defined
[00:01:46.750]by a soil's ability to infiltrate water.
[00:01:50.360]We can't go out there and play on a field,
[00:01:54.190]play sport on a field or use it for a gathering
[00:01:56.680]if it's water logged, if it's just a big mud pit.
[00:02:01.830]So poor water movement through turf soils can really end up
[00:02:05.320]costing a lot of money and unintended consequences.
[00:02:10.870]One quote stating,
[00:02:11.927]"Proper water management is the primary key to success
[00:02:15.750]for most commercial turfgrass facilities,"
[00:02:18.900]is pretty hard to argue with
[00:02:20.400]when you consider all of the factors above.
[00:02:26.240]So we're doing this research
[00:02:29.140]in order to better our understanding
[00:02:31.960]of water movement within the soil profile
[00:02:35.110]and how this water changes over time.
[00:02:40.820]We are going to accomplish this
[00:02:42.510]by developing an improved cost effective method
[00:02:46.090]for tracing water movement that is less invasive,
[00:02:49.335]but reliable and efficient,
[00:02:52.730]but doesn't require extensive technical equipment.
[00:02:58.650]We have set three goals for ourselves to achieve,
[00:03:01.300]and if we can accomplish these goals,
[00:03:04.010]we can consider the project a success.
[00:03:08.070]First we set out to develop
[00:03:10.010]a method that accurately and reliably traces
[00:03:12.560]the flow of water through modified root zones.
[00:03:15.810]This is what I talked a lot about
[00:03:17.360]last year in my presentation.
[00:03:20.350]Second, we plan to develop an algorithm
[00:03:23.000]to derive percolation rate
[00:03:24.380]based on the advancement of the wetting front.
[00:03:28.750]And third,
[00:03:30.240]we want to see if we can achieve
[00:03:32.900]more refined units of millimeters per second
[00:03:36.120]when talking about water movement.
[00:03:37.820]Traditionally, infiltration and subsequent percolation
[00:03:41.417]is characterized by units of centimeters per hour
[00:03:44.850]or even larger units.
[00:03:46.270]But if we can use more refined units,
[00:03:51.970]millimeters per second, we can gain a better understanding
[00:03:54.661]of more refined soil processes such as soil layering
[00:04:00.150]and other things that can inhibit soil water movement.
[00:04:08.830]So our first task was
[00:04:10.170]to develop a method to trace water flow.
[00:04:13.360]We took over this little room in Keim Hall,
[00:04:15.590]not much bigger than a closet and made it our dark room
[00:04:19.191]where we do all of our experiments.
[00:04:22.090]We utilize a green fluorescent dye and black lights
[00:04:25.790]to visualize water flow
[00:04:28.310]and two time lapse cameras to record movement.
[00:04:32.420]Tension table replicates the tension
[00:04:35.350]that occurs at the bottom of golf course putting greens.
[00:04:38.600]And this tension occurs
[00:04:39.680]because of the hanging water column effect of
[00:04:42.440]a fine textured material
[00:04:47.840]overlying a coarser textured material.
[00:04:54.140]Next, we use 12 inch transparent plastic cylinders,
[00:04:58.890]which enable us to see the soil profile and water flow.
[00:05:04.050]And we decided 12 inches
[00:05:05.620]because this is the average depth of a USGA putting green.
[00:05:11.060]Now, prior to sampling,
[00:05:12.660]we have to apply two coats of a hydrophobic polymer coating
[00:05:16.530]to the inside of the cylinders.
[00:05:18.870]This coating repels water from the cylinder walls,
[00:05:24.160]so we are using it in order to promote even infiltration.
[00:05:32.060]This is a soil probe that we've used to take our samples.
[00:05:35.390]It's the most current probe that we have.
[00:05:37.860]The samples that I'll be using as an example
[00:05:40.460]in this presentation were taken with this probe.
[00:05:46.144]And this probe is powered
[00:05:47.340]by a truck jack that is turned upside down
[00:05:51.810]and is driven into the soil with a handheld drill.
[00:05:56.690]Four augers that you can see here in the background
[00:05:59.050]are what keep this soil probe anchored into the ground.
[00:06:03.890]And here is a steel cylinder,
[00:06:07.480]which houses that plastic cylinder
[00:06:13.210]and prevents it from being destroyed during sampling.
[00:06:21.710]This is a picture of what the putting green looks like
[00:06:24.730]after a sample has been removed.
[00:06:27.650]Disturbance is quite minimal.
[00:06:30.070]The augers leave little marks
[00:06:31.960]that are equivalent to bar ball mark repair
[00:06:35.260]and then we fill up the cylinder with sand.
[00:06:46.570]In here are three cores
[00:06:47.700]taken from three different golf courses here in Lincoln,
[00:06:51.290]Highlands, Mahoney and Hillcrest.
[00:06:56.360]And I do wanna point out that these Hillcrest samples
[00:06:58.700]look like there'll be really interesting to sample,
[00:07:02.490]to do experiments on
[00:07:04.130]because if we can see about halfway down this profile
[00:07:07.060]it turns to mostly soil based,
[00:07:10.379]it's not heavily sand anymore.
[00:07:12.934]And these profiles are actually pushup greens,
[00:07:17.890]so they utilize that native soil
[00:07:19.933]that's already there in construction
[00:07:23.510]and they date back all the way to 1928,
[00:07:26.160]almost a hundred years old.
[00:07:28.310]So it will be really interesting
[00:07:29.970]to see how water moves through these profiles.
[00:07:32.960]I haven't run any of them yet,
[00:07:34.280]but I have a feeling they're gonna be very interesting.
[00:07:38.800]So now that we have our intact soil core,
[00:07:41.490]we have to saturate it for 24 hours prior to our experiment.
[00:07:46.500]After the core has saturated,
[00:07:48.210]we can put it on the tension table
[00:07:50.670]where immediately a bentonite clay slurry is applied.
[00:07:55.480]And this bentonite clay is simply
[00:07:57.730]clay and water mixed together
[00:07:59.490]and it is applied to the interface
[00:08:02.670]between the cylinder wall and the soil core.
[00:08:06.116]And this is done to prevent preferential edge flow,
[00:08:09.220]which produces a flow pattern
[00:08:10.740]that is not representative of the soil
[00:08:12.830]from which we took our sample.
[00:08:19.260]So now that we have our intact soil core,
[00:08:24.080]we are able to apply the fluorescent dye
[00:08:26.852]and record water movement.
[00:08:31.730]From this video,
[00:08:32.940]we can see pretty quick moving wetting front
[00:08:37.920]that diffuses evenly down the core.
[00:08:41.110]There doesn't appear to be
[00:08:42.500]much obstruction of water movement,
[00:08:46.960]and infiltration of our two inches of applied dye
[00:08:50.930]eventually ceases, and this completes our first goal.
[00:08:59.880]We can check off our first realization
[00:09:02.860]of developing a method that accurately and reliably traces
[00:09:06.650]flow of water through modified root zones.
[00:09:09.420]Now on to our second goal.
[00:09:15.960]Once we have taken a recording of water movement,
[00:09:18.680]we can use image processing software
[00:09:20.850]to derive percolation rate
[00:09:22.530]as water advances down the soil core.
[00:09:25.710]We've been using a program called MATLAB
[00:09:27.810]to create our algorithm.
[00:09:30.420]And we've been getting a lot of help from Yufeng Ge,
[00:09:33.080]the Biological Systems Engineering Department
[00:09:35.900]with creating and developing this algorithm.
[00:09:45.180]So, one of the first things we do
[00:09:47.860]is isolate our area of interest.
[00:09:50.490]We define our area of interest as the soil core.
[00:09:54.010]So we're basically telling the program
[00:09:55.560]to ignore everything outside this purple-pink box.
[00:10:01.110]We also have to create this blue line,
[00:10:05.640]which follows the advancement of the wetting front
[00:10:08.520]and is what records are data points
[00:10:10.640]that we will use to calculate percolation rate.
[00:10:14.950]So I play this video
[00:10:18.170]and we can see the program tracking the wetting front.
[00:10:21.220]Ideally the blue line
[00:10:22.530]should move smoothly down the soil core,
[00:10:25.610]unlike in this video
[00:10:26.860]where we see the line oscillating and jumping around a lot.
[00:10:31.060]And this is due to a lot of noise
[00:10:33.130]that is created in the video
[00:10:35.360]and it's something that we've been working on to reduce.
[00:10:38.888]We've been working with improving the lighting,
[00:10:42.680]as well as different angles
[00:10:44.770]to see if this can improve the image quality.
[00:10:50.020]But like I said, this blue line represents the wetting front
[00:10:53.580]and it is what records the data that we need.
[00:11:00.690]Now, we can graph this raw data
[00:11:04.480]but it doesn't really tell us much.
[00:11:06.100]It shows us somewhat of a trend,
[00:11:07.770]we can see where infiltration starts initially,
[00:11:11.380]where the line starts moving up and down
[00:11:13.700]and we can also see towards the end here
[00:11:16.240]where infiltration starts to cease, becomes constant.
[00:11:21.600]And this graph just basically plots
[00:11:23.460]the amount of fluorescent pixels in each frame.
[00:11:28.160]And as might be expected from the video we just watched,
[00:11:32.320]just as in that video is in this graph
[00:11:35.240]there's a lot of noise,
[00:11:36.660]this really should be a smooth line.
[00:11:38.890]So we have to apply a smoothing filter
[00:11:41.150]to help reduce this noise.
[00:11:46.810]The filter we applied to this data is called Savitzy-Golay
[00:11:50.670]and it significantly reduces the noise
[00:11:52.711]and leaves us with data that we can use
[00:11:54.820]to calculate percolation rate.
[00:11:59.230]Now, this graph is created knowing the length of the core
[00:12:03.270]and the amount of pixels in this length,
[00:12:08.240]which is used to calculate the length of each pixel.
[00:12:12.700]The change in pixels from frame to frame
[00:12:15.700]represents length of downward diffusion.
[00:12:19.100]Because we know the video speed is 30 frames per second
[00:12:23.820]we can calculate speed of this downward diffusion,
[00:12:28.040]and this graph is our final product.
[00:12:30.880]This graph shows water movement
[00:12:32.680]that percolated very uniformly which can be expected.
[00:12:37.740]When we remember the video of infiltration,
[00:12:40.960]it showed a quick uniform infiltration
[00:12:45.330]that was fairly uninhibited.
[00:12:48.930]We can see that percolation rate
[00:12:50.830]is greatest at initial infiltration into the soil surface,
[00:12:55.070]which corresponds to the rapid filling of flow paths
[00:12:58.030]as the water enters the soil profile,
[00:13:01.650]so this is what we would expect.
[00:13:05.370]We move down to the bottom of the profile,
[00:13:09.950]around 18 centimeters,
[00:13:13.290]we can see that percolation starts to flat line,
[00:13:15.980]it starts to become constant and then ceases altogether.
[00:13:22.750]This also validates our video where we didn't see the dye
[00:13:26.160]fully reach the bottom of the core,
[00:13:28.920]it only reached about eight inches.
[00:13:32.610]Now using this graph,
[00:13:34.370]we can go along at any point, at any depth
[00:13:38.250]and see how fast water was percolating at that depth.
[00:13:49.760]So we have accomplished our goal of developing an algorithm
[00:13:54.500]and deriving percolation rate.
[00:13:58.040]Now we have to ask if,
[00:14:01.090]did we realize our measurements in millimeters per second?
[00:14:05.670]And yes, we indeed did.
[00:14:09.070]So with these measurements,
[00:14:11.410]we can start to distinguish different soil layers.
[00:14:15.780]There really doesn't appear
[00:14:16.760]to be any restrictive layering development
[00:14:19.690]in the video that we examined today.
[00:14:24.800]We can start to understand
[00:14:26.320]how different management practices
[00:14:28.030]like top dressing and soil aeration
[00:14:30.700]influence layer development.
[00:14:34.600]We can then use this information to provide evidence
[00:14:38.690]of how we manage turfgrass and turfgrass soils
[00:14:43.320]impacts water movement.
[00:14:46.959]And we do all this in the hopes of being able
[00:14:49.490]to improve management recommendations
[00:14:52.120]to ensure proper water movement.
[00:14:57.550]Now our current work,
[00:14:58.500]we are currently having a new probe being built
[00:15:02.450]by the Programming, Instrumentation and Electronics Group
[00:15:06.160]which is part of
[00:15:06.993]the Biological Systems Engineering Department.
[00:15:10.920]This new probe will have a motor
[00:15:13.810]to evenly insert the cylinder into the soil profile
[00:15:18.930]with much more uniformity
[00:15:23.680]than can be applied using a handheld drill.
[00:15:27.210]So this will reduce the disturbance to the soil core
[00:15:30.920]and improve the quality of the sample.
[00:15:34.530]Second, we are adding four sensing equipment
[00:15:37.430]that record the force being applied to the soil core
[00:15:42.450]as it moves with depth.
[00:15:45.350]This will provide a way to validate our percolation rate
[00:15:48.240]we calculated from our videos
[00:15:51.010]and further help the understanding
[00:15:53.340]of the presence of soil layers within the profile.
[00:15:58.960]So we are very excited to have this soil probe built
[00:16:03.320]and test it out to see all the possibilities
[00:16:07.300]so if it will really improve our project,
[00:16:08.989]improve the quality.
[00:16:12.360]So here's a picture that we were just sent
[00:16:15.990]of what our new probe will look like.
[00:16:18.770]You can still see the space for the steel cylinder
[00:16:21.760]to house our transparent plastic core,
[00:16:24.630]four places for the augers to hold down.
[00:16:26.980]It looks like they added some wheels
[00:16:28.450]onto the back for easier towing.
[00:16:31.990]And, then we can see on top all of our fancy equipment
[00:16:35.290]that they've added to smooth our actuation method
[00:16:38.300]and record the force sensing.
[00:16:42.390]So yeah, I'm really excited to see what all this can do.
[00:16:45.890]Should be done within the next couple of days, definitely.
[00:16:48.732]Hopefully, by the end of next week.
[00:16:51.140]So this is what we'll be working on
[00:16:53.150]through the fall and into the spring as I finish up.
[00:16:58.770]I just wanna say thank you again for joining me
[00:17:01.470]and have a great day.
[00:17:03.556](upbeat music)

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Water Movement in Turfgrass Soils

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments