Field sampling:Samples were collected from the Bradley Street Pitch Pine Stand (North Haven, CT) during the morning of October 20, 1998. Because we wish to replicate their methods as closely as possible, the techniques used in 1990 by Anne Marsh were demonstrated by Thomas Siccama, an investigator from her experiment, who then aided and observed the sampling.
At each of five different sample pits chosen randomly throughout the site, a 15 x 15 square template guided the extraction of a block of forest floor. After removing the forest floor, a stainless steel soil coring tube (8.4 cm in diameter) was plunged vertically down into the mineral soil to a depth of at least 20 cm such as to be below the bottom of the Ap. The insertion was usually done in one smooth rapid motion so as to minimize perturbation of soil layers, although light hammering was used if the tube did not descend to a depth of 20 cm. The surrounding soil was removed, and the tube was carefully eased to a horizontal orientation so as not to mix the soil along the open slit or to allow some columns of soil to fall out of the tube. The excess soil was gently removed such that the soil was flat and even with the edges of the slit. Measurements were marked on the soil face starting from the uppermost horizon to divide the soil into slices of increasing thickness down to 20 cm:
Depth - cm 0-2 2-4 4-6 6-8 8-10 10-13 13-16 16-20 This differential slice thickness was based on previous experiments showing concentration changing less rapidly at greater depths. The soil was then pushed slowly out of the tube from the top, and divisions were made at the measurement points so as to have volumetric samples according to their depth. By marking the places to cut off slices before extruding the core, effects of compression are minimized since the narks were made before pushing on the core. Only one site contained any substantial presence of roots in the soil slices. Samples were placed in ziplock bags for return to the laboratory.
Processing in the lab:
The forest floors were oven dried (80 oC), weighed and then ground in a Wiley Mill to pass a 2 mm stainless steel screen. Mineral soil samples were oven dried (80 oC) to constant weight. Mineral soil was sifted through a 2-mm stainless steel sieve to remove the coarse fraction. Coarse fractions were divided into rocks and roots and the weight of the rocks, roots, and oven-dried soil were then taken separately.
Samples from the same forest taken by Marsh and Siccama in 1990 were available. Thus we were not dependent on published or previous analyses in making our comparisons. Samples from 1990 and 1998 were processed and analyzed at the same point in time with the same equipment.
Two grams of each sample, including 1990 samples and 1998 samples, were then dry-ashed for at least 8 hours at 500 oC in a muffle furnace. The ash was weighed, eluted with 6 N nitric acid, and diluted to 50 mL with nanopure water. An empty crucible (process blank) and a sample of apple leaves (standard reference material) were processed with each batch of 12 samples processed.
The aqueous solutions were analyzed for concentrations of 11 elements using a Perkin-Elmer Optima 3000 Inductively Coupled Plasma Atomic Emission Spectrometer.
Data for forest floors from 1980 were not reanalyzed; the data was taken from previous results derived by Mark Groff in 1980-1. For lead, zinc, and copper, these results were later reanalyzed, with a close correlation. This similarity gives some testament to the accuracy of these results for comparative purposes.
Go back to Front Page
IntroductionGo on to
Results