Summer Institute Group Projects, 2013

Group Projects:


Ocean Health Evaluation: The utility of the cumulative
impact index in the Southern California Bight

As humans are expanding current and planned uses of the ocean, it is increasingly important to understand how these impacts affect coastal ecosystems.  A cumulative human impact score was created for the California Current by synthesizing fine-scale information on the distribution of multiple environmental stressors (Halpern et al. 2009).  The resulting cumulative impact score is potentially useful to managers and policy makers, because it narrows a complex interaction of factors into a single number that can directly contribute to environmental management decisions.  These scores take into account multiple aspects of human impacts, such as climate change, pollution, and fishing pressures; however, it remains unclear if impact scores reflect the ecological status of marine systems. 


The goal of this project is to examine the relationship of the cumulative human impact score with ecosystem conditions (e.g., biodiversity, community structure, and stability), focusing on the Southern California Bight and subtidal kelp forest ecosystems.  We will use abundance and biomass data collected by PISCO and LTER between 1999 and 2011 for kelp forest communities.  We will analyze the relationship of the overall cumulative impact index and subsets of it with several community indices that reflect ecosystem status and potentially identify thresholds for these communities with differing levels of human pressures.  These analyses will provide an evaluation of how well impact scores reflect the status of kelp forest ecosystems and improve their potential use in the policy and management of coastal resources.


Dispersal Biogeography 

Under shifting climates, many species will not be able to survive in their current locations. To avoid extinction, these species need to track the shifting climate. Species mobility is thus expected to critically influence species’ ability to persist. Dispersal models that use species traits to understand how far species can move often ignore the geographic context. Conversely, habitat models rarely accommodate relevant dispersal traits. Realistic predictions about whether dispersing species will be able to track their suitable habitat, despite potential barriers like mountains and rivers, will require an integrated approach.

In our NCEAS Summer Institute project, we use relevant functional traits, including seed type and growth form, to derive a suite of mathematical models of species-specific dispersal. These models estimate the probability of species dispersal to a given distance. We then place the calculated probabilities into a spatial context to infer whether species will keep pace with their suitable habitat. We also investigate whether certain groups of species are more likely to track shifting climate, and whether these species share functional traits. If they do, we can identify how climate will affect the structure and composition of plant communities around the globe.

Figure above: Changes in the functional richness, entropy and dispersion resulting
from the potential extinction of species sensitive to climate change. All three measures of
functional diversity are expected to decline. Randomizations show that the decline is not significantly
different from that expected under random extinctions. These preliminary results suggest that climate
change-induced extinctions are not selective and affect all species regardless of their ecology.



Trends in river nitrogen, phosphorus, and sediment
in the agricultural Midwest US, and the relation to
federal expenditures towards conservation practices 

Elevated loads of nitrogen (N), phosphorus (P), and sediment in rivers present an obstacle to achieving sustainable agriculture, water resources, and fisheries. Losses of N, P, and soil from cropland, as well as wildlife concerns, have prompted a large national effort to develop and implement conservation practices through local, state, and federal partnerships with landowners. As a consequence, several federally supported conservation practices thought to benefit water quality are now widespread throughout the Midwestern US. Many of these practices have been implemented through programs of the US Department of Agriculture, and cumulative federal support of conservation practices since 1990 totals in the tens of billions of $US (inflation-corrected 2010 dollars). Despite widespread adoption of some conservation practices, annual N and P loads of the Mississippi River and other large agricultural rivers of the US show no substantial changes since the year 2000, and loads have actually increased in some areas. But now, longer water quality records, more river monitoring sites, and finer resolution for conservation spending are available, offer new opportunities to identify conservation effects needed to inform land management decisions.

On our project we are asking: What is the relationship between conservation spending and water quality trends for streams and rivers of the Midwest? To answer this question, we will explore the relationship between conservation spending and water quality trends from the scale of individual basins to large regions for several hundred streams and rivers of the Midwestern US. Our analysis will also consider other basin-scale factors that influence water quality, such as the use of tile drains, trends in precipitation, agricultural intensity, and the presence of urban areas.


Taxonomic Diversity and Ecosystem Function

VIMS Marine Biodiversity Lab manager Paul Richardson stocks
experimental mesocosms in the York River Estuary.


The widespread and increasingly rapid extinction of species on planet Earth is a well-documented phenomenon, and two decades of empirical research manipulating species richness has shown that biodiversity loss has important consequences for the structure and functioning of ecosystems. However, in a large number of experiments, the relationship between diversity and functioning is saturating, not linear, suggesting that only a few species are necessary to sustain even moderate levels of functioning. Why is this? One possibility stems from the fact that when we design experiments, we treat each species as if it could have an independent effect on functioning. Yet, we know that certain species are more similar than others in terms of their morphology and physiology, and consequently, those species are likely to be redundant in terms of their effect on functioning. Species that share morphological or physiological traits are often also closely related, suggesting that the degree of shared evolutionary history within an assemblage may be a suitable proxy for functional redundancy. Indeed, preliminary work in terrestrial grasslands has shown that the phylogenetic diversity of a community is a better predictor of primary producer biomass than is simple species counts. It remains unclear, however, whether this finding can be generalized to other systems and taxa.

Our group is using a database of existing experimental manipulations of species richness to test whether the effect of diversity changes based on the degree of evolutionary relatedness within the experimental assemblage. To answer this question, we have assembled taxonomic information on more than 700 species from nearly 500 experiments, and will use this information to calculate an index of taxonomic diversity. For a subset of these species, we have also collected gene sequence information with which to construct gene trees, from which we will calculate an index of phylogenetic diversity. We expect the effect of diversity on functioning will be stronger when assemblages are more taxonomically or phylogenetically diverse. Moving forward, investigations of the relationship between biodiversity and ecosystem functioning will require an explicit consideration of the degree of functional redundancy among species when generating experimental assemblages.


The Urban Biodiversity Project 

In this synthesis, we examine the relative impact of environmental and socioeconomic factors on urban biodiversity patterns across the United States.  Our ultimate objective is to reveal how specific local and regional land use, policy, and socioeconomic factors could be managed for increased urban biodiversity.  Specifically, we are using citizen science records, global biodiversity databases, city-level attributes, national land cover, and census data to examine how the diversity and community composition of birds, mammals, and pollinators varies across ~500 cities in the US.  We also plan to examine functional changes in community composition across sites in order to identify species and functional groups most at risk to intensified urbanization.  Our analytic approach involves the combination of ordination and generalized linear mixed models in order to identify the most important predictors of species richness and community composition across the three taxa. We will identify cities that support greater wildlife diversity than expected given its landscape or climate, and investigate which city policies may be contributing to their success.  This will be among the first studies to focus on fine-scale (e.g., 0.5km-2.5km) environmental and socioeconomic factors across a wide range of US cities, and will complement existing work in the field by simultaneously examining bird, mammal, and insect biodiversity.  This synthetic analysis will engage citizen scientists and enable city planners to enact policies that encourage biodiversity.