5.4 How the Data Will Be Gathered, Stored, and Used
sentative tropical forests (and in other Biosphere
Reserve areas worldwide) to inventory and monitor changes in biological diversity (Dallmeier,
1992; Stohlgren, 1999). An extensive commitment
to data management is evident: the program employs a full-time data manager, and the Smithsonian's own BIOMON software program (1.
Comiskey, pers. comm.) was developed to store,
analyze, and archive information. Other programs
with an extensive information management program include the U.S. Department of Agriculture
Forest Service Forest Health Monitoring Program
(Burkman and Hertel, 1992) and Forest Response
Program (Zedaker and Nicholas, 1990) and the U.S.
Environmental Protection Agency's Environmental Monitoring and Assessment Program (EMAP;
Messer et aI., 1991; Palmer et aI., 1991).
The challenge of obtaining and maintaining
long-term support for data acquisition cannot be
overstated (Stohlgren et aI., 1995b). Funding
sources appear more and more reluctant to support
long-term projects, yet this support is necessary for
continued consistent measurements and for maintaining data management systems and organizational infrastructures (Strayer et aI., 1986; Franklin
et aI., 1990). Government agencies and others are
downsizing the permanent research staff necessary
to maintain continuity of field measurements over
time and relying more on temporary staff. Regardless of the difficulties involved, the true measure
of success in ecological studies should be the successful preservation and long-term use of the data.
5.4 How the Data Will Be
Gathered, Stored, and Used
Throughout an ecological assessment, and especially in the initiation phase, investigators need a
detailed vision of how the data will be gathered,
stored, analyzed, and used (Risser and Treworgy,
1986). This involves careful planning, vision, and
a strong commitment to information management.
The planning phase includes the development of a
detailed study plan (usually derived from the study
proposal) with concomitant peer reviews and statistical reviews. The vision phase includes elaborate foresight of data and information products
(e.g., data sets, GIS themes, summaries, graphics,
mathematical models, and publications) from the
data gathering stage to the analysis stage (discussed
next). However, plans and vision are rendered useless without the obligatory commitment to information management.
73
5.4.1 Gathering Data
Although remotely sensed data are common, I will
limit specific examples to data collected in the
field. For example, in many forest monitoring studies, each tree is mapped, tagged, and identified, and
the plots are recensused annually (Dallmeier et aI.,
1992). The 50-ha permanent plot on Barro Colorado Island, Panama (Hubbell and Foster, 1987),
the North American Sugar Maple Decline Project
(Miller et aI., 1991), and the USDA Forest Service's monitoring programs continue to accrue and
manage detailed data sets (Zedaker and Nicholas,
1990; Burkman and Hertel, 1992). It may be prudent to seek the advice of those with years of experience.
Several new tools are available to field ecologists to improve the efficiency of data collection.
Global positioning systems help locate and relocate
plots in rugged terrain and rangeland expanses.
Electronic range finders improve the efficiency and
accuracy of tree mapping. Palm-top computers are
replacing data sheets, permitting immediate data
entry and analysis. In the office, a GIS often provides the central framework for data management.
A GIS can combine maps of biotic and abiotic information with historic data and new data. Perhaps
the most underutilized capabilities of a GIS are in
the unbiased selection of assessment and monitoring sites (Stohlgren and Bachand, 1997) and the
predictive ecological modeling to support management decisions (e.g., Buckley et aI., 1993). Spatially explicit, high-resolution GIS data are available from satellites and aerial photography.
However, acquiring and using remotely sensed data
require decisions of scale and resolution, the number and type of sensors, and whether the data are
georeferenced or not (see Section 5.5).
5.4.2 Storing and Archiving Data
Data storage depends largely on the hardware and
software used and the particular format desired. It
is wise to store data sets in two different buildings,
in two formats (one of them being ASCII), and on
two media (e.g., tape and disks). The most important aspects of metadata (data describing data) are
data set name, responsible organization, contact
person and address, project codes, data form and
format, key words, geographic coverage, scale,
time span of data, accessibility, and additional comments about the data sets. This includes spatially
explicit data that are incorporated into a GIS. Metadata should follow, as closely as possible, the Fed-
sentative tropical forests (and in other Biosphere
Reserve areas worldwide) to inventory and monitor changes in biological diversity (Dallmeier,
1992; Stohlgren, 1999). An extensive commitment
to data management is evident: the program employs a full-time data manager, and the Smithsonian's own BIOMON software program (1.
Comiskey, pers. comm.) was developed to store,
analyze, and archive information. Other programs
with an extensive information management program include the U.S. Department of Agriculture
Forest Service Forest Health Monitoring Program
(Burkman and Hertel, 1992) and Forest Response
Program (Zedaker and Nicholas, 1990) and the U.S.
Environmental Protection Agency's Environmental Monitoring and Assessment Program (EMAP;
Messer et aI., 1991; Palmer et aI., 1991).
The challenge of obtaining and maintaining
long-term support for data acquisition cannot be
overstated (Stohlgren et aI., 1995b). Funding
sources appear more and more reluctant to support
long-term projects, yet this support is necessary for
continued consistent measurements and for maintaining data management systems and organizational infrastructures (Strayer et aI., 1986; Franklin
et aI., 1990). Government agencies and others are
downsizing the permanent research staff necessary
to maintain continuity of field measurements over
time and relying more on temporary staff. Regardless of the difficulties involved, the true measure
of success in ecological studies should be the successful preservation and long-term use of the data.
5.4 How the Data Will Be
Gathered, Stored, and Used
Throughout an ecological assessment, and especially in the initiation phase, investigators need a
detailed vision of how the data will be gathered,
stored, analyzed, and used (Risser and Treworgy,
1986). This involves careful planning, vision, and
a strong commitment to information management.
The planning phase includes the development of a
detailed study plan (usually derived from the study
proposal) with concomitant peer reviews and statistical reviews. The vision phase includes elaborate foresight of data and information products
(e.g., data sets, GIS themes, summaries, graphics,
mathematical models, and publications) from the
data gathering stage to the analysis stage (discussed
next). However, plans and vision are rendered useless without the obligatory commitment to information management.
73
5.4.1 Gathering Data
Although remotely sensed data are common, I will
limit specific examples to data collected in the
field. For example, in many forest monitoring studies, each tree is mapped, tagged, and identified, and
the plots are recensused annually (Dallmeier et aI.,
1992). The 50-ha permanent plot on Barro Colorado Island, Panama (Hubbell and Foster, 1987),
the North American Sugar Maple Decline Project
(Miller et aI., 1991), and the USDA Forest Service's monitoring programs continue to accrue and
manage detailed data sets (Zedaker and Nicholas,
1990; Burkman and Hertel, 1992). It may be prudent to seek the advice of those with years of experience.
Several new tools are available to field ecologists to improve the efficiency of data collection.
Global positioning systems help locate and relocate
plots in rugged terrain and rangeland expanses.
Electronic range finders improve the efficiency and
accuracy of tree mapping. Palm-top computers are
replacing data sheets, permitting immediate data
entry and analysis. In the office, a GIS often provides the central framework for data management.
A GIS can combine maps of biotic and abiotic information with historic data and new data. Perhaps
the most underutilized capabilities of a GIS are in
the unbiased selection of assessment and monitoring sites (Stohlgren and Bachand, 1997) and the
predictive ecological modeling to support management decisions (e.g., Buckley et aI., 1993). Spatially explicit, high-resolution GIS data are available from satellites and aerial photography.
However, acquiring and using remotely sensed data
require decisions of scale and resolution, the number and type of sensors, and whether the data are
georeferenced or not (see Section 5.5).
5.4.2 Storing and Archiving Data
Data storage depends largely on the hardware and
software used and the particular format desired. It
is wise to store data sets in two different buildings,
in two formats (one of them being ASCII), and on
two media (e.g., tape and disks). The most important aspects of metadata (data describing data) are
data set name, responsible organization, contact
person and address, project codes, data form and
format, key words, geographic coverage, scale,
time span of data, accessibility, and additional comments about the data sets. This includes spatially
explicit data that are incorporated into a GIS. Metadata should follow, as closely as possible, the Fed-