General:
This data set is part of a study including three on-going plot-level fertilization sites.  The first (LTE-MP-LPA) is in the low marsh at Law's Point along the Rowley River.  The second (LTE-MP-LPP) is in the high marsh at Law's Point along the Rowley River. The third (LTE-MP-ORT) is in a brackish water marsh on an upstream tributary of the Parker River.  Biomass sampling is slightly different at each site, and is accomplished non-destructively at the LTE-MP-LPA and LTE-MP-ORT sites, and destructively at the LTE-MP-LPP site.  Biomass sampling at LTE-ME-ORTwas discontinued after August 2008.

At each site a system of boardwalks was built to incorporate 12 experimental plots (1m2 each).  There are 4 treatments (three replicates each) within the boardwalk structure.  The 4 treatments are Control, N-fertilized, P-fertilized, N+P fertilized.  Additionally, at each site there are 3 control plots and 3 N+P fertilization plots external to the boardwalk structure.  These external plots contain porewater samplers, and, in some cases, biomass census plots. 

Fertilizer is buried at several locations within each treated plot, monthly, from May-Sept.  From 1999-2001 the annual fertilization rate was 30 molN/m2/y and/or 15 mol P/m2/y.  From 2002 to present fertilization rate is 7.5 mol N/m2/y and/or 3.75 mol P/m2/y.  N fertilizer was NH4NO3 through 2008, and currently is nitrate/urea/sulfate (34-0-0).  P fertilizer is P2O5 (0-44-0).

Sampling Description:
The plant census method at LPA is non-destructive.  Two permanent subplots were established in each plot within the boardwalk.  Each subplot is 10cmx10cm.  The subplots are labeled 1&2, with subplot 1 being the more northerly and/or westerly subplot.  In May and June, plants in the subplots are tagged with bird ID bands.  Approximately monthly from May to October, heights of the tagged plants and associated culms are measured.  Each plant is stretched upward, and the distance from the sediment surface to the tip of the tallest leaf is measured in cm.  Presence of inflorescences is recorded, but the inflorescence is not included in the plant height.  Plant heights are read into a digital recorder.  Descriptions of the locations of the plants within the quadrat are also recorded (e.g. northwest corner or northwest of tag #xxxx).  Plant measurements are always taken with the observer on the same side of the quadrat for ease in tracking  individual plants throughout the growing season.  Each plant in a subplot is givena unique ID (subplot_plant_id), determined by the birdband ID and location of the plant in relationship to the banded plant.  The same plant ID may be used in more than one subplot.

The plant census at LPA began in June 1999.  All 24 subplots were censused until October 2003.  Beginning in May 2004, only the 6 Control subplots were censused.  In May 2006 the Control census plots (and porewater samplers) were relocated outside of the boardwalk structure because of potential contaminatin from fertilized plots.  The census subplots kept the same id#’s from when they were within the boardwalk for continuity within the ACCESS database.

Data processing:
Plant data are entered first into an EXCEL spreadsheet, and eventually transferred to an ACCESS data base.  Outputs from ACCESS are processed with two FORTRAN programs.  The FORTRAN programs use algorithms to convert the ACCESS data outputs to plant weight, density, births, deaths and growth (Morris and Haskin, 1990).  Fortran outputs are entered into SAS programs to calculate treatment means of biomass, density, and productivity.

References:
(Morris and Haskin, 1990. Ecology 71(6):2209-2217)
(SAS System for Windows; release 8.02; SAS Institute, Inc; Cary, NC)