In November of 1998 four rural, forested plots were established at
Oregon Ridge Park in Baltimore County northeast of the Gwynns Falls
Watershed. Oregon Ridge Park contains Pond Branch, the forested
reference watershed for BES. Two of these four plots are located on
the top of a slope; the other two are located midway up the slope.
In June of 2010 measurements at the mid-slope sites on Pond Branch
were discontinued. Monuments and equipment remain at the two plots.
These plots were replaced with two lowland riparian plots; Oregon
upper riparian and Oregon lower riparian. Each riparian sites has four
5 cm by 1-2.5 meter depth slotted wells laid perpendicular to the
stream, four tension lysimeters at 10 cm depth, five time domain
reflectometry probes, and four trace gas flux chambers in the two
dominant microtopographic features of the riparian zones - high spots
(hummocks) and low spots (hollows).
Four urban, forested plots were established in November 1998, two at
Leakin Park and two adjacent to Hillsdale Park in west Baltimore City
in the Gwynns Falls. One of the plots in Hillsdale Park was abandoned
in 2004 due to continued vandalism.
In May 1999 two grass, lawn plots were established at McDonogh School
in Baltimore County west of the city in the Gwynns Falls. One of these
plots is an extremely low intensity management area (mowed once or
twice a year) and one is in a low intensity management area (frequent
mowing, no fertilizer or herbicide use). In 2009, the McDonogh plots
were abandoned due to management changes at the school.
Two grass lawn plots were established on the campus of the University
of Maryland, Baltimore County (UMBC) in fall 2000. One of these plots
is in a medium intensity management area (frequent mowing, moderate
applications of fertilizer and herbicides) and one is in a high
intensity management area (frequent mowing, high applications of
fertilizer and herbicides).
Trace gas (N2O, CO2, and CH4) fluxes from soil to the atmosphere were
measured every four to six weeks at each plot using an in situ chamber
design. The lawn plots had chambers identical to those used by Bowden
et al. (1990) installed flush with the soil surface to allow mowing.
These (three per plot, at least 5 m apart and at least 5 m from the
plot boundary) consisted of 28.7-cm diameter (ID) by 4.0-cm high
polyvinyl chloride (PVC) cylinders which were placed on permanently
installed PVC base rings immediately prior to measurement.
From 1998 through September 2007, the forest plots had the chambers
described by Goldman et al. (1995), constructed from 16.5 cm wide by
20 cm long pieces of PVC pipe fitted with a septum and an air-tight
well cap. These chambers were placed 4 cm into the soil and had a
total volume of 2 L. Beginning in September 2007, the Bowden et al.
(1990) chambers, which are used at many research sites across the
world, were installed at the forest sites. A brief period of side by
side measurements suggested that the Bowden chambers produce lower gas
fluxes than the Goldman chambers. The replacement dates for chambers
at different sites are:
Leakin 1 September 2007
Leakin 2 September 2007
Hillsdale 1 Sept 2007
Oregon Ridge Upper 1 October 2007
Oregon Ridge Upper 2 October 2007
Oregon Ridge Middle 1 September 2007
Oregon Ridge Middle 2 September 2007
At 0, 10, 20 and 30 min following placement of the chamber on the base
or installation of the well cap, 9-mL gas samples were collected from
gas sampling ports in the center of the chamber top by syringe.
Samples were injected into 9 mL evacuated glass vials sealed with
septa and aluminum rings. Samples were sent to the Cary Institute of
Ecosystem Studies (CIES) for gas chromatography (GC) analysis.
Concentrations of CO2 and N2O are measured on a GC equipped with
thermal conductivity and electron capture detectors aligned in
sequence. Concentrations of CH4 are measured on a GC equipped with a
flame ionization detector. Samples and standards are introduced into
this GC via syringe and sample loop. Fluxes were calculated from the
linear rate of change in gas concentration, the chamber internal
volume and soil surface area.
In May 2016 sample collection and analysis change. Samples are now
collected 5.9 mL exetainers (Labco Ltd, UK). 8 ml of samples were
injected into each vial. Analysis is now performed on a GC equipped
with thermal conductivity, electron capture, and flame ionization
detectors (GC2014, Shimadzu Inc.)
Spike standards are processed along with each batch of flux samples.
These standards are generated by filling sample vials from a standard
tank in the field laboratory in Baltimore and then processing these
samples along with samples taken from flux chambers in the field.
Percent recoveries of the trip standards are used to correct flux
values for leakage or gas absorption that may occur between sampling
and analysis. Exceptionally low recoveries for N2O (traced to a change
in septa manufacture) led to loss of flux data for this gas for
several dates from 2007 - 2009.
Literature Cited
Bowden R, Steudler P, Melillo J and Aber J. 1990. Annual nitrous oxide
fluxes from temperate forest soils in the northeastern United States.
J. Geophys. Res.Atmos. 95, 13997 14005.
Goldman, M. B., P. M. Groffman, R. V. Pouyat, M. J. McDonnell, and S.
T. A. Pickett. 1995. CH4 uptake and N availability in forest soils
along an urban to rural gradient. Soil Biology and Biochemistry
27:281-286.
