In 1991, an irrigation transect experiment was established near the Konza Prairie HQ to assess the effects of supplemental water on ecological processes in tallgrass prairie. The site is burned annually in the spring. The transect spans upland, hillside and lowland topographic positions with irrigation and sampling points (12) located at 10 m intervals. Adjacent control transects are marked on both sides of the irrigation transect. Irrigation is scheduled according to estimates of actual evapotranspiration and measures of plant water status. In 1992, an additional 4 irrigation sprinklers were added to the transect (2 at each end). In 1993, a second line of sprinklers and control plots was added (#16-31). At the time of peak aboveground biomass (late August to October), six 0.1 m2 quadrats are harvested at each of the 30 sites (no #9 due to rock outcrop) for the irrigated and control/non-irrigated lines. Biomass is separated into live grass, forb and woody. As of 2006, c.dead is no longer separated from live grass. Vegetative species composition was initially measured in 1991 at each site, and continues to be measured at midseason by using a modified Daubenmire canopy coverage technique in a 10 m2 circular plot. At approximately 10 day intervals, predawn and midday plant water potentials are measured in Andropogon gerardii at each site in both irrigated and control transects. Since 1992, reproductive effort of the dominant grasses Andropogon gerardii (ANGE), Sorghastrum nutans (SONU), Schizachyrium scoparius (ANSC) has been assessed in irrigated and control transects by measuring heights (n=9) and densities (n=4) of flowering stalks. In 1993, soil moisture measurements at 15 and 30 cm depths were begun with a Time Domain Reflectometry system. Data set also includes measured natural precipitation and supplemental water added to the site.
Full data used in the Broderick et al. (2022) paper, "Climate legacies determine grassland responses to future rainfall regimes", can be found at the link: http://lter.konza.ksu.edu/content/wat02-climate-legacies-determine-grass....
To assess the long-term response of selected vegetation and ecosystem parameters to annual burning with no water limitation. To determine long-term changes in canopy cover, richness, and diversity of plant species in irrigated and non-irrigated uplands and lowlands.
Location of sampling: The transect and two control transects are located about 300 m (600 ft) southwest of the old stone reservoir and windmill to the west-southwest of headquarters (grid B-16). The transect runs perpendicular to the slope just south of the belowground plots on the east side of the drainage way. The area is identifiable by the line of sprinklers on 1 m risers from the 7.5 cm diameter aluminum pipe which runs a length of 140 m (460 ft) down the transect (Map available as following).
The area which is subject to additional water is outlined by a row of steel posts located 15 m (50 ft) on each side of the line of sprinklers. Posts are located 18 m (60 ft) apart along the transect to correspond to every other sprinkler on the line. A numbering system is laid out in reference to the sprinklers. The lower most sprinkler is 1 and the upper most sprinkler is 15. This pattern is repeated for sprinklers 16-31. Every other sprinkler has a number mounted on its riser. The steel posts have a number, which corresponds to the number on the sprinkler it is nearest. The total research area consists of the area inside the steel posts plus how much farther outside the potential wetted area the particular experiment needs to extend to get sufficient unirrigated area to compare to the irrigated area. As research areas outside the areas bounded by the steel posts is requested, individual researchers will be responsible to mark those areas with steel posts.
Walking on the area is necessary to make measurements, get samples, etc. When walking to a particular area, please walk either parallel or perpendicular to the sprinkler line and line of steel posts. When walking up the transect or perpendicular to the transect, please walk on lines directly between steel posts or lines formed by steel posts and the closest sprinkler. Researchers may walk on their plots as necessary. Walk only on the lines between steel posts and sprinklers to get to individual research areas.
When an area is assigned to a researcher it will be located with reference to the nearest sprinkler, the irrigation line, and the row of steel posts. You may mark it as you see fit.
Species Composition Plots: Species composition conduits are located east of each riser in the irrigated transect and in the eastern control transect. Plant composition plots are marked with a conduit in transects that are parallel to the irrigation sprinklers. Another transect that is parallel to the irrigated plots but outside the reach of the supplemental water is used as control plots. TDR probes for soil moisture measurements are located west of the sprinkler risers in the irrigation transect and in the western control transect. Biomass is harvested on the north side of the risers away from the TDR probes, and reproductive effort is assessed on the eastern side of the irrigation transect and in the eastern control transect. The non-irrigated samples are collected in a random circle 2 m away from the species composition marker. Reproductive effort is done similar. Leaves for plant xylem pressure potential measurements are collected from either side of the transect.
Sampling history: Plant compositions plots were established in 1991. Initially, 13 irrigated and 13 non-irrigated plots were sampled, but in 1992 the number of plots expanded to 30 in both treatments.
Frequency of sampling: Aboveground biomass is sampled in late August-October. Reproductive effort is sampled in October. Plant water potential predawn and midday (1300 CDT) and soil moisture is sampled at ca. weekly to 10-day intervals depending on the weather. Soil chemistry is sampled at five-year intervals. All plots are sampled for species composition once each year in late July.
Variables measured: 1. Aboveground biomass; 2. Plant reproductive effort; 3. Xylem pressure potential in Andropogon gerardii; 4. Plant species composition; 5. Soil moisture and chemistry.
Canopy cover of all vascular plant species in each plot are estimated. Because sampling is only conducted in the summer, the cover of early-spring plants and cool-season grasses are likely underestimated.
Irrigation Procedures: General Information: A single line of full-circle sprinklers with pressure regulators below each one are spaced close together on the supply line to provide relatively uniform amount of water at locations parallel to the line. There is a gradient, in terms of water added, from most water at the line to none at the outer radius of the sprinklers (15 m from the line with no wind interference). Scheduling of irrigation will be done according to the needs of the plants along the transect near the irrigation line. Plant stress will be monitored as the measure of when to irrigate. Nominal supplemental water needs for grass in the Manhattan area averages about 450 mm per year for cool season types. Warm season grasses will probably require less because their active growing season is shorter then for cool season types. During the growing season from June through early September, however, water use should be similar for both types of grasses, which is about 350 mm on an average year. Nominal water use from all sources is likely to be about 6 mm per day for a fully watered condition. A nominal irrigation provides about 20 mm of water along the sprinkler line. So, two irrigations per week will be needed if no rain occurs.
A well at the reservoir provides about 12 gpm into the reservoir. An irrigation pump takes water from the reservoir and delivers it to the sprinklers at about 90 gpm. A nominal irrigation will be to run for three hours. During irrigation with the well running, the reservoir drops about 1 foot per hour. So, the reservoir, which has about 3.5 feet of working depth when full, must be within about 0.5 feet of the top (at least 3 feet on the scale in the reservoir) before a three hour irrigation can begin.
Preparing for Irrigation: First, the well must be started far enough in advance of the time for irrigation to get the reservoir full. The fill rate will vary with the condition of the well. A nominal fill rate is 0.2 feet per hour. So, for an irrigation that begins at 9 am, and the reservoir is reading 1 foot on the scale, the well should be started at least 14 hours before irrigation is to begin to 7 pm the day before. No harm is done if the reservoir overflows. Electricity, however, is wasted and the future capacity of the well may be jeopardized if a lot of water is wasted. The well pump is controlled by the switch in the electrical control box located about 30 feet east of the windmill. It is the one with the 20 on it (top one of the two). As soon as an irrigation is finished, the well pump should be turned off. You may see some water outside the reservoir when the tank is nearly full. There is a small leak somewhere near the top of the reservoir that has not been located. The reservoir can sit full with no concern, so filling can be done as convenient.
The 24 raingauges within the transect between Sprinklers 3 and 4 must be empty before you begin irrigating.
Performing the Irrigation: Water should be applied when the wind is less than 10 mph (16 kph). Early morning or late evening is usually the best time, however, this does not fit well with effective use of classified staff because it involves overtime. Experience has shown that by avoiding conditions with winds in excess of 20 mph provides acceptable distribution of water on the transect. Check the local forecast as you plan your irrigation. Once you begin, it is recommended that irrigation continue even though the wind is uncooperative. Nominal watering time is three hours. The raingauge network will provide an estimate of water applied across the transect. We assume that the distribution is similar for other locations along the transect. To carry out an irrigation do the following:
Amount of water applied: The amount of water applied to each part of the transect will be determined from the grid of raingauges between Sprinklers 3 and 4. It will be assumed that the distribution at other locations along the transect is similar to this location. This information will include accumulated amount of additional water referenced to the distance from the sprinkler line. The expected distribution is to have about 350-400 mm more added near the line and none out 15 m from the line. The gradient will not be exactly uniform because of wind and the nature of the sprinklers. The expected typical gradient of additional water as a percent of the amount 2 m from the line of sprinklers is shown in Appendix M.
Plant species composition (WAT012):
Due to the new treatments on the irrigation transect, we need to change the way we handle the plots in our datasets. Recall that there are 60 plots (30 irrigated, 30 control) in 2 replicate transects (plots 1-15 in transect A, 16-31 in transect B). Historically, we have simply listed the plot number and treatment to designate plots, but the treatment reversal will cause a considerable amount of confusion under this system. Therefore, we have decided to divide the plots into 4 new transects (combining previous transect and previous treatment into a single variable).
Aboveground biomass (WAT013):
Methods are identical to those in data set PAB011 except six 0.1 m2 quadrats are harvested at each sampling point /sprinkler. (AK) All irrigated samples are taken on the north side of the sprinkler within 4 ft of the spigot to ensure highest possible water content (approximately 100%). Non-irrigated/control samples are collected in a random circle 2 m away from the species composition marker. Due to rock outcrop, no collections are made at sprinkler no.9. Summary of changes: Only four 0.1m sq quadrants were collected in 1991. #26 irrigated location has a large amount of poison ivy in the vicinity. Samples collected from both sides of sprinkler instead of only north side.
Every year since 2003, #8 (irrigated) has had a large dead zone around it due to chemical removal of B. bladhii. Measurements are either taken outside the 'normal' range or not at all (2007). A few other locations have had small spots (approximately 1 ft in diameter) from chemical removal, specific notes are made on the data sheets. These dead spots are not clipped but may force the collector to go further away from the normal collection area.
As of 1999, only samples A, B, and C from plots 1,2,3,13,14,15,16,17,18,29,30, and 31 (irrigated and control) are kept for further analysis.1, 2, 3, 16, 17, and 18 are lowland.13, 14, 15, 29, 30, and 31 are upland. Starting with 2006 samples, 'c. dead' is no longer separated from 'live grass'. Categories are live, forb and woody.
2017: samples A, B, C from 4, 5, 6, 10, 11, 12, 19, 20, 21, 26, 27 and 28 irrigated and non-irrigated will now be kept for further analysis.
Plant reproductive effort (WAT015):
We estimate density of flowering culms of Andropogon gerardii (ANGE), Sorghastrum nutans (SONU), and Schizachyrium scoparium (ANSC) by counting all reproductive culms in four randomly placed 0.25 m2 quadrats at each sampling point. Heights of reproductive culms are measured to the nearest cm by selecting the nearest culms of each species at nine (previous manual states 3 but it has always been nine) randomly selected points at each sampling location. No harvesting occurs in this sampling scheme. Due to rock outcrop at sprinkler #9, no measurements are taken.
Summary of changes for reproductive effort (WAT015): Every year since 2003, #8 (irrigated) has had a large dead zone around it due to chemical removal of B. bladhii. Measurements are taken outside the normal range. Other areas have had small kill zones; specific notes are made on data sheetes. These 'spots' did not impact collections: generally these spots are less than 1 ft in diameter.
Xylem pressure potential: At least seven mature leaf blades are collected at each sampling location and immediately stored in a plastic bag with wet filter paper. Leaves are transported back to the lab and xylem pressure potential is measured with a PMS Model 1000 Pressure Chamber to the nearest 0.1 MPa (1 bar). At the limestone break along the irrigation and control transect, leaves from a hackberry tree are also collected and measured.
Soil Moisture (WAT017):
Thirty six Campbell Scientific CS616 probes, or Time Domain Relectrometry (TDR) probes were installed in the Irrigation Transect study site early August of 2017 to evaluate soil moisture content across two topographic gradients with and without water limitations. Half of the sensors were positioned in an upland site (plot# 1-18) located along transects C and D, and the remaining eighteen sensors were positioned in a lowland site (plot# 37-54) along transects A and B. Probes were inserted in the ground at a depth of 15 cm next to a designated 1 m² plant species composition plot. Each topographic position is instrumented with a Campbell Scientific CR1000 Datalogger that is programed to collect volumetric water content every one half hour. Data is collected in the field and archived on the network server.
For List of Species Codes used, please check: species list (PDF), or species list in Excel version
For the WAT Experiment Map, please check: WAT map
For additional metadata information see: http://lter.konza.ksu.edu/sites/default/files/DC.pdf
For additional methods information see: http://lter.konza.ksu.edu/sites/default/files/MM.pdf
11/28/2017: WAT012- Due to the new treatments on the irrigation transect, we need to change the way we handle the plots in our datasets. Recall that there are 60 plots (30 irrigated, 30 control) in 2 replicate transects (plots 1-15 in transect A, 16-31 in transect B). Historically, we have simply listed the plot number and treatment to designate plots, but the treatment reversal will cause a considerable amount of confusion under this system. Therefore, we have decided to divide the plots into 4 new transects (combining previous transect and previous treatment into a single variable).