Revegetation and Succession of Western Kansas Riparian Site

PROJECT BACKGROUND
Our family farming operation is a typical western Kansas crop and livestock farm. Our most important crop is wheat, with 450-500 acres planted each year. Wheat is both summer-fallowed and continuous-cropped, depending on rainfall. Fall crops (250-300 acre) include grain sorghum, sunflowers, and sudan grass. The farm also has about 40 acres of alfalfa. Weed control is achieved by chemicals and cultivation. Crop rotation is practiced on almost all ground except continuous wheat. Continuous wheat is only grown 2 years in a row before summer fallowing. The farm also includes cattle and hogs. Hogs are confinement grown. Hog manure is stored in an old oil collection tanks until spread on cultivated land. Cattle are rotated between pastures both during years and between years, depending on cattle numbers and rainfall. No special sustainable practices other than terracing erodible ground have been part of this operation.

PROJECT DESCRIPTION AND RESULTS
My goal for this project was to demonstrate the feasibility of restoring natural riparian vegetation in a near climax form for water quality improvement, wildlife protection, and streambed stabilization.

Site preparation: the site was divided into 3 treatment plots – grazing, no grazing, and no grazing with vegetation enhancement. Plots were burned in early April to encourage nutrient recycling and grass growth.

Experimental procedure: several tests were developed to scientifically measure variation between plots. These were:
- Diversity: Step-point transect data was collected twice. The plant nearest to the point was identified. In addition, a plant collection was made at the beginning of the season and pressed for later identification. Diversity was used to suggest the condition of the site.
- Cover: Step-point transect data was collected twice. Each point was identified as hitting or missing a plant. Percent cover could be figured from this data. Cover was also a measure of site condition and plant productivity.
- Density: A forage metering disk was used along a transect twice to measure above ground plant material. The disk was calibrated the first time it was used. Density suggested plant productivity/given area.
- Insect population: Insects were trapped in each plot by means of an alcynite cylinder covered with sticky paper. Counts were made weekly and paper was changed. Insects were assumed to be an indicator of wildlife food available, especially for birds.
- Water quality: Water samples were taken from the streambed at the beginning of the experiment. Phosphorus levels were measured as an indicator of animal contamination. No end-of-season collection as possible as the stream had dried up.

Project Assistance:
- Fort Hays State University: Dr. Robert Nicholson and Dr. Thomas Wenke provided technical assistance with experimental design and phosphorus testing.
- Wildlife and Parks: Daryl Fisher made recommendations on plantings for the enhanced vegetation plot, burning advice, and provided shrubs for the plots. He also suggested that I apply for funds from wildlife groups.
- Pheasants Forever: Arlen Ricke provided me with forms and procedures for obtaining cost-share money. Money was used for plastic-type ground cover along shrub rows.
- Kansas Co-operative Extension: the letter required in applying for this SARE grant was written by the Ness county extension agent Scott Barrows. Jim Collins, an agent from an adjoining county, spoke at the meeting. The Ness county extension council also provided all equipment for the meeting. Materials for the meeting were ordered from the central office in Manhattan.
- SARE: Ken Schneider visited the project and suggested ways to improve it. He also provided encouragement and support for what I was doing.

Results: Results have not been analyzed at this time. However, review of the data suggests these conclusions:
- Phosphorus levels were so low that water quality is probably not a problem on this site
- Insect counts dropped precipitously after a record heat wave in August. Stable flies, which had made up over half the collection in early counts, became more minor component later in the season. Fly counts and total number of insects were highest on the grazed plot. The vegetation-enhanced plot had the most diversity of insects.
- Wildlife observations were made as insect counts were collected. As an inexperience observer, probably I missed many signs. Blackbirds were common in the spring. Traces of frogs, deer, pheasants, raccoons, and some burrowing animals were evident.
- Diversity of plant materials was greatest on the grazed plot. This can be explained in part by the fact that there is the greatest ground variation within this plot, including a section of limestone breaks. Trends in diversity cannot be measured in a single year.
- While not officially part of the experiment, grazing patterns where observed. Big bluestem in the grazed plot was actively selected by cattle. Patches of big bluestem which extended into the adjoining pasture were grazed much more heavily than surrounding grasses. Cottonwoods, which emerged along the bank in all plots, were grazed by cattle but not killed. Cattle trampled some of the stream bank, but stocking rate (8 acres/cow-calf) was moderate enough that no serious damage was done. Also, water sat longest in the part of the stream bed in the grazed plot. This kept the soil damp for a longer period, allowing more evidence of trampling.
- Vegetation quantity was not widely variable between plots. It would be expected that any resulting differences between plots would take several years to be observed.
- Basal cover, like vegetation quantity, did not differ greatly between plots. Cover was greater in the spring than in the fall on all plots.

My general observations on the project are that stream bed restoration is possible for the average producer. Grass growth was excellent due to heavy rains this spring. Tall grasses exceeded 5’ in height. The most important factors in stream bed restoration would be burning at the appropriate time in the spring and minimizing the impact of cattle. Ideally, cattle would be excluded form the immediate vicinity of the stream bed. The major drawback is the cost of fencing out the stream. Acreage lost is minimal. Grasses are perceived as being of exceptional quality along the stream, but I believe that very little forage is lost as a result of fencing. Fencing out a section of a pasture also gives an idea of the potential of the site.

The biggest failure was the shrub and tree planting. Despite the weed barrier, survival was not very good. An extremely wet, late spring delayed planting, and late summer was exceptionally hot and dry. I expect those shrubs which survived to eventually spread. The loss of trees was offset by the unexpected appearance of native cottonwoods. While not numerous (I would guess less than thirty), cottonwoods can spread rapidly and colonize a stream bank. I was surprised at how rapidly cottonwoods moved into the treatment area. Upstream, in an ungrazed section of the creek, there are mature cottonwoods and a weedy under story. I walked part of this section in the spring and did not find a single young cottonwood. I presume burning the plots made a more congenial site for seedling cottonwoods.

The treatment plots affected our overall farming operation very little. Cattle used for the treatment were out of sync with the rest of the herd due to late calving and needed to be kept separate anyway. Fencing the plots out of the pasture was a major task, but existing fence was of very poor quality and needed replacing anyway. The grazed plot was fenced with electric wire. I would suggest that as fence is replaced on a farm, streambed protection should be considered as one of the criteria, along with water development and cross-fencing for grazing efficiency.

I do not expect any great impact from this small section of riparian restoration. However, such restoration can be used many places and each bit helps the whole. It is relatively easy and low tech to do what I did, well within the reach of any producer. The number of people who asked to hunt indicates that the treated area was perceived as being better for wildlife than average pasture drainage. Fencing out an acre of land per half mile of creek (my estimate) works out to less than $400/quarter-section in land costs in this part of the country. Since hunters easily spend half this much yearly on hunting blind leases, it does not seem exorbitant. Taxes on this amount of land are minimal.

OUTREACH
The major public outreach associated with this project was a producer meeting held this fall. At the meeting, each agency that assisted with the project spoke about their contribution and about their area of expertise. I presented the scope of the project; the procedures used, and showed slides of the results. Attendance was very poor (less than 10) despite substantial advertising in two local weekly papers and numerous signs posted in 5 area towns. I have submitted a proposal for a poster presentation at the annual meeting of Society for Range Management next February. I am awaiting evaluation.