Target species: Q. douglasii
Life stages: mature trees
Origin: natural
Situation: range
Locations: Sierra Foothill Research and Extension Center, Yuba County
Summary: This study looked at how trees and low intensity grazing affect nutrient pools and soil physical properties in a blue oak woodland. Compared to adjacent grasslands, soils beneath oak canopy had a lower bulk density, higher pH, and greater concentrations of organic carbon, nitrogen, total and available phosphorus, and exchangeable calcium, magnesium and potassium, especially in the upper soil horizon. The authors suggest that oak trees create islands of enhanced fertility beneath their canopy through organic matter incorporation and nutrient cycling. As noted below, grazing was not adequately addressed as a treatment factor, so no valid conclusions about grazing effects can be drawn from this study.
Methods: Four "treatments" were examined: oak with grazing, oak without grazing, grassland (no canopy) with grazing and grassland without grazing. Soil pits were excavated to bedrock (80-120 cm) with a backhoe either beneath oak canopy or at least 3 meters from the edge of oak canopy. Five soil pits within each grazing/canopy combination were sampled. The authors treated their data as a split plot design (main plot=grazing treatment, subplot=canopy). However, the grazing treatment was not randomly assigned because the nongrazed area is essentially a single large exclosure. Therefore, the grazing treatment is not truly replicated. The "replicates" are actual subsamples (a.k.a. pseudoreplicates) of a single experimental unit, so the split-plot model is not statistically valid. Using such a model, one cannot validly test for effects of grazing. The comparisons made between grazed and nongrazed fields at best are comparisons of two specific fields; there is no statistical basis to attribute the differences (or lack thereof) to grazing per se. The use of multiple comparison procedures to compare the four "treatment" combinations is also not valid. Although the reported significance levels of detected differences may be inaccurate, overall comparisons between canopied and uncanopied areas appear to be valid.
The nongrazed area is a formerly-grazed parcel from which cattle have been excluded for 20 years prior to the study. Depending on its past grazing history and other factors, soil conditions in this area may still reflect past grazing impacts, as the authors note. Futhermore, grazing intensity in the grazed area was quite light compared to typical commercial grazing practices; for the 20 years prior to the study, beef cattle were reportedly stocked at a rate of approximately 0.5 animals per hectare for 4 to 8 weeks between January and June (Dahlgren and Singer 1991 report a somewhat different grazing regime for the same parcel). The low intensity of grazing is illustrated by the fact that peak standing grass biomass was actually slightly greater in the grazed field than in the nongrazed field. Hence, even if a statistically valid design was used, one would still anticipate that the differences between the grazed and nongrazed areas would be minimized due to the selection of experimental units.
Because the date(s) that the soil pits were dug are not reported, the temporal relationship between this paper and the related Dahlgren and Singer 1991 paper is unclear.
Differences in soil properties between fields ("grazing effects"). As noted above, due to a lack of replication, differences noted between grazed and nongrazed fields cannot be attributed to grazing alone. All such differences could be associated with sampling error and other factors that vary spatially. Overall, few differences were noted between the fields. In the nongrazed field, the A soil horizons were thicker and soil bulk density lower than in the grazed field, differences that could be due to soil compaction and destruction of A horizon by cattle. The grazed field also had slightly lower pH in the A horizon and higher available phosphorus than the nongrazed field. The authors speculate that cattle dung and urine were the source of the additional available P in grazed areas, especially under oak trees where cattle congregate during hot weather. This explanation is intuitive and plausible alrhough not rigorously documented.
Differences in soil properties related to canopy. Compared with noncanopied
areas, soil under oak trees showed the following differences:
thicker A horizon;
lower bulk density;
soil pH higher in the upper soil horizons and lower in the deeper part of
the soil profile;
higher concentrations of carbon, nitrogen, and total and available phosphorus
in all soil horizons;
higher amounts of organic carbon and nitrogen in the soil solution;
higher cation exchange capacity;
higher levels of exchangeable calcium, magnesium, and potassium concentrations
in the upper soil horizons.
The authors related most of these differences to the greater amounts of nutrient cycling associated with oak litterfall beneath canopy. The amount of organic matter returned to the soil surface beneath oak canopy was nearly four times that returned in the open grasslands. The authors also note that leaching of nutrients is reduced under oaks because of their greater rooting depth compared to grasses, greater drying of the soil due to increased transpiration, and interception by canopy of water reaching the soil surface.
This study is related to Dahlgren et al 1997 and was conducted at the same location. Bulk precipitation, canopy throughfall, and soil solution samples were collected from January to June, 1990. These were analyzed for pH and various nutrient cations and anions.
Differences between fields ("grazing effects"). The grazed -nongrazed comparison suffers from the same design limitation described for Dahlgren et al 1997, hence differences ascribed to grazing are not statisically valid. The major differences seen between grazed and nongrazed fields were higher levels of soil sodium (Na) and chloride (Cl) in the soil solution of the grazed field. The authors attribute higher Na and Cl levels in the grazed areas to input from the salt supplement provided to the cattle, which is plausible though not demonstrated.
Differences related to canopy. Especially in the A horizon, soil solutions from under oak canopy had higher nutrient levels (phosphate, ammonium, sulfate, calcium, magnesium, potassium) and higher pH than those from noncanopied grassland. Rainfall dripping through the canopies of Q. douglasii and Q. wislizeni trees showed elevated levels of potassium, calcium, magnesium, chloride, phosphate, ammonium, and pH relative to rainfall.
The authors speculate that phosphorus is the most limiting nutrient for plant growth at the site, but in Dahlgren et al 1997 they conclude available phosphorus is actually present at levels in excess of plant uptake requirements.
We have not located this report for review.