Geoscience Reference
In-Depth Information
Imaging LiDAR is especially powerful for tracking changes in surface
elevation, above-ground vegetation biomass, 3D vegetation structure, and 3D
distribution of canopy-leaf area. Particularly in forested regions, LiDAR can be
used to improve estimates of net ecosystem production and carbon stocks over
large areas (Goetz and Dubayah 2011; Hall et al. 2011). Imaging radar has also
been an important tool for monitoring vegetation structure, especially in cloud-
prone areas and areas subjected to seasonal flooding (Bergen et al. 2009). Image
fusion, the combined use of imagery from two or more sensors, can be used to
exploit complementary information from very-high-resolution multispectral
imagery, hyperspectral imagery, and LiDAR (Koetz et al. 2007).
Long-Term Datasets in Real-Time
Dense, long-term environmental datasets in real-time could create a foun-
dation for informed decision-making. A suite of decision-support tools could be
developed for integration with air- and water-quality models at various scales.
For example, data in hospital admission forms could be combined with meteoro-
logical and air quality models in real time to provide health forecasts and warn-
ings. Real-time sensing and modeling of water-borne pathogens in situ could
provide drinking water treatment plants with threat forecasts, alerting them to
the need to change source water or treatment techniques. Special research atten-
tion could be given to handling uncertainty of both data and models. In addition
to deterministic models with uncertainty analyses, probabilistic approaches can
be extremely powerful when computational intelligence tools are used.
Data assimilation and data mining approaches provides innovative possi-
bilities. An example is the use of an intelligent real-time cyberinfrastructure-
based information system called the Intelligent Digital Watershed to better un-
derstand the interactions and dynamics between human activity and water qual-
ity and quantity. Such an approach provides “1) novel uses of data mining algo-
rithms in data quality and model construction, 2) development of specialized
data mining algorithms for [environmental forecasting] applications, 3) devel-
opment of data transformation algorithms, [4)] data-driven modeling of non-
stationary processes, [such as storm forecasting for by-pass wastewater dis-
charges], and [5)] development of decision-making algorithms for models con-
structed with data mining algorithms”.
4
Using data in a novel way could greatly
expand the analysis capability of EPA and provide insights previously impossi-
ble to obtain without such innovations.
Already, the Consortium of Universities for the Advancement of Hydro-
logic Science, Inc. Hydrologic Information System (HIS) project has a system-
atic data acquisition network for the publication, discovery, and access of water
4
NSF-CDI. 2008-2011. CDI-Type II: Understanding Water-Human Dynamics with
Intelligent Digital Watersheds. (#0835607). Jerald L. Schnoor (PI), David Bennett, An-
drew Kusiak, Marian Muste, and Silvia Secchi.
Search WWH ::
Custom Search