Abstract
Military landscapes represent a mixture of
undisturbed natural ecosystems, developed areas, and lands
that support different types and intensities of military
training. Research to understand water-quality influences of
military landscapes usually involves intensive sampling in a
few watersheds. In this study, we developed a survey design
of accessible headwater watersheds intended to improve our
ability to distinguish land–water relationships in general,
and training influences, in particular, on Fort Stewart, GA.
We sampled and analyzed water from watershed outlets.
We successfully developed correlative models for total
suspended solids (TSS), total nitrogen (TN), organic carbon
(OC), and organic nitrogen (ON), which dominated in this
blackwater ecosystem. TSS tended to be greater in samples
after rainfall and during the growing season, and models
that included %Wetland suggested a ‘‘build-and-flush’’
relationship. We also detected a positive association
between TSS and tank-training, which suggests a need to
intercept sediment-laden runoff from training areas. Models
for OC showed a negative association with %Grassland.
TN and ON both showed negative associations with
%Grassland, %Wetland, and %Forest. Unexpected positive
associations were observed between OC and equipmenttraining
activity and between ON and %Bare ground ?
Roads. Future studies that combine our survey-based
approach with more intensive monitoring of the timing and
intensity of training would be needed to better understand
the mechanisms for these empirical relationships involving
military training. Looking beyond local effects on Fort
Stewart streams, we explore questions about how exports of
OC and nitrogen from coastal military installations ultimately
influence estuaries downstream.
