Final Report for GS03-020
Project Information
Our overall research objectives were to assess the differences between innovative soybean tillage systems (no-till) and traditional soybean systems (till) for ecological impacts on pen-raised bobwhites. Chapters of this thesis were divided into three main research areas: (1) population estimate and flush count survey; (2) insect abundance and soybean measurements; and, (3) liberation studies for pen-raised bobwhites.
Bobwhite densities were relatively low on Clemson University¡¦s Pee Dee Research and Education Center, with the highest mean density estimate of 0.048 bobwhites per hectare found surrounding Field B. Density estimates as great as 6.2 bobwhites per hectare have been reported at Tall Timbers Research Station, Leon County, FL, but a density estimate of 2.5 bobwhites per hectare may be more realistic as a maximum density supported an ideal habitat in South Carolina (Wellendorf 2002). A difference in maximum density (carrying capacity) between South Carolina and northern Florida is primarily believed to be due to variation in native vegetation and soil characteristics.
Flush counts produced limited data for assessing differences between habitat types because wild bobwhite population densities were low, causing a reduced detection probability which limits sample sizes for statistical analysis. Our data suggest that this method of sampling should not be conducted on areas that have density estimates below 1.00 bobwhites per hectare.
No-till soybean systems were found to result in faster canopy closure and have a higher insect abundance than both the tilled soybean fields and forested areas for the most beneficial Orders OrthoƒVptera, AraƒWchnida, and ColeoƒWptera/HemiƒVptera. Faster canopy closure with no-till systems was mostly due to the narrow row widths. No-till soybean fields also had more insects per trapping event than field borders for araƒWchnids and coleoƒWpterans and hemiƒVpterans, but not for orthoƒWpterans. Increases in canopy closure and insect abundance from narrow row widths, no-till farming practices, and are respectively are directly related to bobwhite fitness and habitat suitability. Faster and greater canopy closure may have provided added protection from avian predators, which allows bobwhites the capability of utilizing these habitats for nest construction, incubation, and brood rearing activities such as foraging. Insects provide both metabolic energy and protein requirements (both adult hen and chicks) for growth and reproduction during this period characterized as high in stress, due to the predator vulnerability of flightless chicks and energy expenditures of reproductively active hens.
Liberated pen-raised bobwhites were found to select no-till treatments (soybean and corn combined) more than tilled treatments, field borders, and forested areas for 5 out of 9 fields during release date two and three. Only Field E had higher bobwhite occurrences within tilled than no-till treatments. This may be related to the proximity of a five to six year-old clear cut, the most ideal habitat surrounding only the tilled treatments and an earlier planting period of two weeks for tilled than the no-till soybean field (same split-plot).
No-till agricultural practices incorporating narrow row widths and transgenic varieties should be more beneficial to bobwhites because these systems could allow for earlier and multiple nesting attempts (Puckett et al. 1995), increased insect abundance (Palmer 1995 and this study) and greater protection for avian predators. These benefits are primarily due to faster canopy closure and greater amounts of soil surface residue left after planting. This form of tillage practice also provides other environmental benefits including: reduced soil erosion, improvements in soil moisture, nutrient-retention of soils, increases in beneficial invertebrates, reduction in fossil fuel usage, less sedimentation of local water bodies, less chemical runoff, reduction in carbon dioxide emission, and as previously stated wildlife enhancement. Our data suggest that no-till systems can possibly aid in the recovery of bobwhite populations, but only if a wildlife habitat management is integrated throughout agricultural fields including: prescribed fire, forested thinning and field border management (to include ditch banks and non-agricultural areas).
Literature Cited
Palmer, W.E. 1995. Effects of modern pesticide and farming systems on northern bobwhite quail brood ecology. Dissertation. North Carolina State University, Raleigh, NC. 131 pp.
Puckett, K.M., W.E. Palmer, P.T. Bromley, J.R. Anderson, T.L. Sharpe. 1995. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 49:505-515.
Wellendorf, S.D., W.E. Palmer, and P.T. Bromley. 2002. Habitat selection of northern bobwhite coveys on two intensive agricultural landscapes in eastern North Carolina. Proceeding of the National Quail Symposium 5:191pp.
Introduction
Many farmers in the Southeastern USA have been switching from traditional tillage practices to more recently developed conservation-tillage systems that include narrow row widths and transgenic varieties. Planting herbicide-tolerant varieties in narrow rows should result in fewer weed problems in conservation tillage systems, were cultivation is generally not used for weed control. Implementing these weed-control practices in conservation tillage systems should also result in the use of more environmentally friendly herbicides, in addition to reducing soil erosion and improving soil productivity.
New production practices must be evaluated not only for their effects on pest populations, but non-targeted species as well. Besides causing shifts in weed and foliar insect populations, new weed-control practices can affect insects living on the soil surface and, hence, the wildlife that feed upon them. Over the last several decades, bobwhite quail (Colinus virginianus) populations have declined in the Southeastern USA, at least partially due to the use of clean tillage practices by farmers. Since quail habitat includes agricultural fields, we conducted a two-year study to examine the affects of new weed-management systems (centering on conservation tillage) on bobwhite quail food availability and habitat quality. An innovative weed-control system for use with conservation tillage was compared to a traditional production system on five split fields at the Pee Dee Research and Education Center in 2002 and 2003. Each field was divided into four sampling areas: conversation tillage system, traditional tillage system, transitional zone, and wooded habitat. Year one (Y1) split fields were all planted with doublecropped soybeans (Glycine max L. Merr) and year two (Y2) fields were planted half in doublecropped soybeans and half in corn (Zea maize L). Quail were monitored using flush counts (Y1) and radio telemetry (Y2) to understand the impacts of tillage treatment and crop selection on quail habitat use and mortality. Insects and vegetative samples were also randomly collected within these four areas to determine habitat attributes including crop canopy coverage for brood and escape cover, insect populations for food availability, and tree stand density for comparison among forested areas. All data such as field location, field characteristics, insect trap location, and forest and intermediate sampling were mapped using GPS and transferred into a GIS system for comparison and modeling.
To determine how new weed-control practices for conservation tillage systems affect quail habitat use or avoidance, foraging, and mortality.
Identify how an innovative weed control system centered on conservation tillage, narrow row widths, and herbicide tolerant varieties affects insect populations that quail use as a major food source.
Assess the potential benefits of conservation and traditional tillage systems with respect to pesticide use.
Cooperators
Research
Year one monitoring of bobwhite quail was conducted using flush counts, which involved four individuals walking parallel through the field separated at determined distances specific for each field. Quail observations were low in that year with only 16 birds flushed during the summer (June through August). Therefore, it was decided that quail monitoring in year 2 should be conducted using radio telemetry. Quail were trapped during the winter and early spring of 2003 under guidelines of Clemson’s Animal Use and Care Protocol and trapping permits from South Carolina Department of Natural Resources (SCDNR). Baited funnel traps were constructed and placed near forest edges throughout the Pee Dee Research Center and Education Center. Quail trapping success was not sufficient to answer the objectives of our study. Therefore, pen-raised quail were used for further study and released on the 23rd of July, 5th of August, and the 19th of August. Twelve quail per field (six per side) were fitted with a 5g radio transmitter and released within each treatment field at predetermined release sites. Quail were tracked daily using homing techniques developed by White and Garrot 1993, using a 3-element yagi antenna and an ATS receiver. Bobwhite quail locations were identified according to crop and tillage treatment, border or wooded habitat. When mortality occurred, remains were located and examined to identify the type of predator that caused the mortality.
To assess crop canopy closure within treatment fields, light interception by the crop canopy was measured throughout the growing season on both sides of the fields. Pitfall insect traps were used in both years to determine differences in insect abundance between the treatment fields, border, and wooded habitat. Two pitfall traps were placed randomly in each treatment (no-till, traditional tillage, border, and woods) and collected weekly. Insects were identified to order and to family in the order Orthoptera.
Year one (Y1) data indicated that flush counts were biased due to the assumptions that were required for this type of survey method. These assumptions included observing all quail that are present and that the quail do not move off the study site without being recorded. The result of the survey method did not indicate a habitat (crop or tillage) preference for bobwhite quail (p> 0.05).
A total of 180 pen-raised quail were released and tracked with radio telemetry during the summer of 2003. Quail mortality was greater than expected compared to past research; however, this previous research was conducted in areas intensively managed for quail and releases were later in the year (September to November). The average life span for the quail released in our study was 7.68 days.
The graphs shown below are the average number of times each quail was observed in each of the treatment habitat types. The conservation tillage (CT) and traditional tillage (TT) treatments included both the corn and soybean crops.
Quail were found primarily in the CT treatment (summed over soybeans and corn) for all fields except for field E. Statistics were run with SAS software using PROC GLM with an alpha level of 0.05. In fields A, B, D, and F, quail preferred the CT treatment compared to the TT treatment. Number of quail found in the CT treatment was also much higher than those found in the woods and border treatments, except field E where the differences were not significant.
Tracking data allowed for identification of treatments that resulted in the greatest number of deaths. Chi-squared analysis was used to compare the treatment areas with an alpha level of 0.05. Location in the wooded areas resulted in the greater number of mortalities. The CT and TT treatments were not significantly different; however the CT treatment resulted in a slightly higher number of mortalities (n=28, n=24).
Mammalian predators accounted for 62% of the quail deaths (Tables 1 – 3). In contrast, avian kills accounted for only 29% of the deaths. Quail released on the last date survived the longest, which probably was due to the greater crop canopy development at that time, compared to the other release dates (see below).
To compare canopy coverage of CT soybeans versus TT soybeans, crop light interception readings were taken throughout the growing season in both years. A LiCor light bar was used to collect this data. Licor reading were analyzed using chi-square analysis to compare the slopes of the trends lines shown below between tillage-system treatments. Fields were combined in the analysis step to increase the sample size for each sample data. Percent light transmission was statistically greater in TT treatments than CT treatments for each year separately and when combined over years (p <0.05). These results indicate that crop canopy closure was greater and achieved earlier in the growing season for the CT treatment, thus providing quail much more protection from predators during this time. Insects were collected weekly beginning three weeks after planting of soybean and continuing for a total of six weeks during year one and seven weeks for year two. Insects were rinsed from a field strainer and transferred into storage containers, frozen, and identified at Clemson’s Institute of Environmental Toxicology Center. The majority of the insects collected were from the Order Orthoptera, mainly in the family Gryllidae (crickets). The orders Hymenoptera (ants) and Coleoptera (ground beetles) also accounted for a large percentage of the insects collected in both years. The bar graph listed below is the sum of insects collected for all fields for each treatment and year. To analyze statistically the differences that occurred between treatments and study years, Proc GLM was used to determine the means and a mean separation was preformed using least square differences (LSD) with an alpha level of 0.05. The Order Arachnida (spiders) did not show many differences for comparisons between treatments or years for fields B, C, and D. In field A, the CT year 1 (Y1) treatment was significantly different than all other treatments except the CT year 2 (Y2) treatment. The Family Gryllidae (crickets) was highly variable between fields, but showed the highest means for CT treatment (Y1 and Y2) for all fields except field C where the highest mean was collected in the border treatment. The same pattern of highest means were recorded for the Order Hymenoptera (ants) in the CT treatment for all fields, except in field A, were CT Y2 treatment had the lowest mean for all treatments. In fields B and D, CT Y1 treatment was significantly different than TT Y1treatment. Woods (Y1) and border (Y1) treatments did not show significant differences for tillage treatment; however in Y2, significant differences were found between theses treatments in fields A and C. For the Order Coleoptera (ground beetles), CT Y1 was significantly greater than all other treatments except for field B, were TT Y1 had a greater mean. Y1 tillage treatments were significantly greater than woods and border for all fields, except field D. The CT Y2 treatment means were greater than TT Y2 treatment means for all fields except field C; however, only field D showed a significant difference between these two treatments. The border treatment in field D again showed an increased mean when compared to the other field border treatments, indicating a highly productive area for insects and diversity. The Order Isopoda and Superoder Colobognatha (centipedes and millipedes) were both highly variable due to low means, but showed trends of woods and border treatments being the main areas where these insects could be found. In Y1, fields A, B, and C had significantly greater means for the Order Isopoda in woods and border versus the tillage treatments. Only field A showed similar results for Y2. For the Superorder Colobognatha, mean collection results were low for all fields except in Y2 for Field C.