Vermont RII track-2 idea submissions

Vermont EPSCoR issued a state-wide call to academia, public sector and private sector contacts for new ideas that support research on water, climate change, environment, sustainability on 23 June and due on 10 July.

We received 2 responses.

The first, from Professor Declan McCabe (St. Michael's College Biology) focuses on impacts of climate change, invasive species, and changing land uses on freshwater benthic communities; a regional perspective utilizing data from the New England states.

Rationale: Climatic effects: New England regional climate changes include warmer temperatures, increased precipitation and associated erosion/deposition, later freeze ups; earlier ice out. Impacts on organisms may include distributional and life history changes. Symptoms of temperature changes are most likely to be detected as local extirpations of northerly-distributed taxa for which New England represents the southern range extent. By the same logic we should anticipate these taxa to first disappear at lower elevation sites. Similarly we might expect new occurrences of southerly-distributed taxa for which New England represents the northern range boundary. These southern taxa should be anticipated to first appear in low elevations and taxa previously restricted to low elevations should extend ranges to higher elevations.

Invasive species: New England has been invaded by a substantial number of aquatic invasive species and is on invasion front of several others. New England State agencies have benthic community sample data dating back decades. These data sets in many cases include repeat sampling of locations before and after an aquatic invasive has moved in (Rusty crayfish in the Connecticut river basin for example; zebra mussels in Lake Champlain). This rich data set offers the potential to broaden our knowledge of the impacts of invasive species, reveal community types where invasive species are more likely to establish, and demarcate northern limits of invasive species (Corbicula for example). These northern limits are in turn likely to be influenced by climate change. A meta-analysis approach utilizing standard effect sizes (e.g. Cohen’s d) to measure effects of invasives on several community metrics or individual species of economic or conservation importance would be valuable.

Land-use conversion: Land has been converted to and from agricultural and forestry usage, and to urban usage at various paces across the New England region. Associated changes in soil erosion and deposition in water bodies is likely to have had substantial impacts on benthic communities. These changes would best be detected using long term data sets at multiple sites with comparable and contrasting land-use histories.

Why a regional approach?: Range extensions or extirpations at one or a few local sites are certainly interesting, but do not reveal overall patterns at scales relevant to climate change. Similarly a change in benthic diversity following the introduction of a species to a water body does not on its own demonstrate the impact of that species in our region. Similar changes recorded from multiple long-term data sets across the region would reveal the big-picture changes occurring in our aquatic habitats.

What would be needed: These questions would require a shared digital data set derived from the New England state agencies and academic institutions. Once such a data base was established, it would almost certainly facilitate other scientific questions not considered in this idea piece.

The second, from Professor Xindong Wu (UVM Computer Science), focused on transportation and data mining.

The primary goal of this proposal is to design and develop state-of-the-art data mining and fusion techniques and modeling tools that provide reliable and real-time transportation network management including traffic congestion prediction, incident identification and bridge structural health monitoring. This suite will be achieved by designing and using advanced data gathering, processing and mining tools to estimate the current and future transportation system performance. The resulting modeling tools will be enabled by emerging technologies suitable for implementation by transportation agencies. A secondary goal is to quantify the effect of travel behavior changes and consequent planning and modeling challenges in the use of advanced technologies. Research findings will be disseminated to multidisciplinary academic outlets, transportation agencies, and emergency management and planning organizations.

Specifically, the proposal will probe how data mining paradigms can be utilized to address three critical research problems that are at the heart of transportation operations at the present time:

* An on-line architecture integrating roadway, bridge, traffic, and crash information for data usage, archiving and sharing that will be achieved by partnerships with ConnectVermont.

* Novel data fusion algorithms accounting for heterogeneous data sources and for real-time congestion and bridge structural performance monitoring, and incident detection on freeways and major rural and urban arterials with a certain degree of reliability; and

* A Geographic Information Systems (GIS)-based visualization tool enabled with the proposed data fusion models and also as a function of landscape and meteorological variables.