Long-term data sets are fundamental to understand complex ecological systems and inform management decisions for conservation of biological diversity, also known as biodiversity. Long-term data sets (>=10 years) consist of observational and experimental data that can reveal responses of flora and fauna populations to agents of mortality and disease, and shifting geographic distributions due to global warming and changes in land use and land cover, including fire, droughts, floods, tree mortality, invasive species, and management practices. These datasets help answer where and when biodiversity is changing and at what rate. This provides an objective method to choose between multiple management strategies and biodiversity measures. This symposium will focus on long-term data sets at the ecosystem, species, and genetic scales; their utility to assess spatial and temporal trends across a variety of ecosystems and their applications at global, regional, and local scales.
|An Introduction to Biological Diversity Long-Term Datasets|
|Gregory A. Smith|
|Long-term datasets have a storied history within theoretical and applied studies of biological diversity. From the development of ideas in community assembly, to studies of species diversity across space and time, to investigations of endangered species; data collected over a decade or more provide a unique window into patterns of change. Today, anthropogenic impacts to the environment form a suite of problems that are detrimental for native species: habitat loss and fragmentation, overexploitation, pollution, climate change, and the global re-distribution of non-native species. In concert, these impacts have had significant deleterious effects on global biological diversity. Long-term datasets have now become an irreplacable resouce for the development of management and conservation strategies for the protection of biological diversity. Such data provide a baseline from which patterns of change in genetic, species, and ecosystem diversity can be quantified. As data are collected over time, responses to change can be monitored and placed within the context of place. Data analysis can focus on multiple spatial and temporal scales in order to craft strategies that fit local, regional, and global objectives. Learning from previous examples of long-term studies will allow common methodologies to be developed for new studies, such that new data will be maximally useful for management and conservation. Use of large datasets have also spurred methods for accessing, analysing, and mapping data in ways that inform on-the-ground applications. To that end, support for existing long-term datasets and data collection sites, as well as the establishment of new methods and locations for the collection of data, will benefit both theoretical and applied studies of biodiversity well into the future, adding an important tool to the repertoire of management and conservation professionals.|
|Biometrics for Complex Long-Term Biodiversity Data Sets: Lessons from the Breeding Bird Survey|
|John R. Sauer; William A. Link; James E. Hines|
|Most of our understanding of changes in avian biodiversity in North America is based on analysis of population change from the North American Breeding Bird Survey (BBS). The BBS provides data at spatial scales ranging from individual survey locations to continental, but analyses at all scales are complicated by the need to accommodate detectability issues during sampling and changes in sampling effort over space and time. Over the years of the survey, a variety of statistical methods have been used to estimate species richness and change in biodiversity from BBS data. In conjunction with studies funded by the United States Environmental Protection Agency, methods for estimating species richness based on capture-recapture heterogeneity models and focusing on local species integrity (i.e., estimating the proportion of a regional species pool that occurs at a site) were developed and used to address a variety of ecological hypotheses. In recent years, focus has shifted to estimating change in biodiversity through “State-of-the-Birds” composite estimates of population change of collections of species. These composite change metrics can be implemented at a variety of geographic scales and can be weighted by species-specific abundance data. Recently developed methods for estimating occupancy along BBS routes also offer alternative estimation procedures that avoid restrictive assumptions of heterogeneity models. These approaches to estimation of biodiversity change, in combination with ever-advancing abilities to fit complex models to BBS data, have made the BBS even more important as a fundamental source of avian data for biodiversity studies.|
|USGS Foundational Data Supportingnational Assessments|
|Alexa J. McKerrow; Abigail L. Benson; Julie S. Prior-Magee; Elizabeth Martin; Annie Simpson; Daniel J. Wieferich|
|Science Analytics and Synthesis (SAS) Program’s Biodiversity Science activities focus on landscape-level understanding of terrestrial, aquatic, and marine species and ecosystems, through development, maintenance, and integration of a suite of synthesized data systems used to conduct biodiversity assessments at regional to national scales. Our biodiversity data are derived through our partnerships with regional, national, and international biodiversity organizations and are organized according to global standards. Key datasets supported by SAS include terrestrial vertebrate species habitat distribution and fish distribution models, the National Terrestrial Ecosystems Dataset, and the Protected Areas Database of the U.S. (PAD-US). These are central to national conservation assessments, including the National Gap Analysis Project (GAP) and the National Fish Habitat Assessment. In addition, SAS is the U.S. node and major hub of the Global Biodiversity Information Facility (GBIF) as well as for the Ocean Biogeographic Information System (OBIS). Finally, SAS is responsible for Biodiversity Serving Our Nation (BISON), a US-wide species mapping initiative to aggregate occurrence records for species distribution modeling, with an emphasis on species data collected by federal agencies, and invasive species and pollinator datasets.Here we will describe datasets developed by the SAS Program, the partnerships that support this work, and the results of recent national assessments. Specifically, we will describe the nearly 2000 terrestrial and aquatic species models, the breadth and depths of the PAD-US, the recent innovations in the GAP/LANDFIRE National Terrestrial Ecosystems Database, and current statistics about the species occurrence data found in GBIF, BISON, and the OBIS. USGS is well-known for development and maintenance of long-term datasets. Over the past few decades we have been able to expand these assets to include datasets central to improving our understanding of the biodiversity of the U.S. and the biotic components of the critical zone.|
|Evolution and Applications of Biodiversity Data from the Natureserve Network|
|NatureServe carries on a biodiversity data legacy that began in 1974, when TNC helped establish the South Carolina Natural Heritage Program to collect and manage data about the status and distribution of species and ecosystems of conservation concern. Fifteen years later, Alaska became the 50th natural heritage program. The NatureServe Network is one of the most successful examples of a distributed network of biodiversity inventories.
A common language underlies both the Network’s cohesion and the value of our biodiversity data, which can be integrated across boundaries and jurisdictions so that species and ecosystems can be understood in their rangewide contexts. A rigorous set of field and data management standards and protocols, called natural heritage methodology, it was borne 50 years ago from a simple but revolutionary idea: the ecological diversity of any given landscape can be measured.
NatureServe focuses on foundational data: the taxonomy, distribution, and conservation status of species and ecosystems, emphasizing those that are rare and imperiled. They have enabled national biodiversity assessments. They underpin the North Atlantic Coastal Plain’s designation as a global biodiversity hotpot. NatureServe data are incorporated into policy, such as the 2012 USFS planning rule.
Yet in the 1970’s, global change was not the pervasive force shaping our scientific methods. Change over time is not the focus of heritage methodology. The majority of documented sites of high conservation value have only been visited once – we are not a monitoring network. Maintaining data currency is an enormous challenge for any long-term biodiversity data enterprise.
Increasingly we are using modern computational approaches, remote sensing, and machine learning to create efficiencies in our assessment methods, to leverage our field verified observations, and to integrate our data into global change analyses. These new tools help demonstrate the invaluable contribution of the 50-year history of natural heritage programs.
|Springs Online: A Long-Term Database of Springs and Spring-Dependent Species, and Its Application in the Springsnail Conservation Strategy|
|Jeff Jenness; Larry E. Stevens; Jeri Ledbetter; Alek Mendoza; Andrea Hazelton; Brianna Mann|
|Long-term datasets provide a uniquely valuable means to observe change over time and to identify emerging threats. Springs Online, a database of over 157,000 springs and over 21,000 surveys hosted by the non-profit Springs Stewardship Institute of the Museum of Northern Arizona, is one such source that focuses specifically on spring ecosystems. Accessible through an online interface, Springs Online hosts spring site and survey data going back decades, gathered from hundreds of users ranging from spring ecologists to citizen scientists. Data include spring type and local geomorphology, flow and water chemistry, species observed, multiple measures of spring condition and threat, images, sketchmaps and multiple locational descriptors. In this presentation we will discuss the Springsnail Conservation Strategy, a joint effort by multiple state and federal agencies and NGOs, as one example of how Springs Online is being used to identify long-term threats and conditions of 100 springsnail species in Nevada and Utah. We are able to quickly map the range of each species, calculate how long it has been since a species was observed at a spring, identify springs at which the most recent survey did not observe the species, and record the history of water chemistry and temperature measurements taken at the springs. For example, a few springsnail species, such as Pyrgulopsis gibba and P. kolobensis, are widely distributed across the southwest, while 38 species are restricted to three or fewer spring locations and thus are highly vulnerable to habitat degradation or loss. Nineteen of these 38 species were last recorded prior to 2001. We will also discuss the difficulty inherent in incorporating datasets from multiple researchers who all use different survey protocols, and strategies we have taken to overcome this problem.|
|Importance of Long-Term Datasets for Research and Conservation of Endangered Species: Kirtland’s Warbler Dataset|
|Carol Bocetti; Deahn Donner; Nathan Cooper; Sarah Rockwell; David Ewert; Joseph Wunderle|
|The Kirtland’s warbler was listed as a federally endangered species from 1967 until 2019. This neotropical migrant is dependent on young, dense, Jack Pine forests on sandy soil with high cover of woody-stemmed ground vegetation on the breeding grounds, and also young, dense coppice with fruit-bearing shrubs on the wintering grounds. This early succession habitat specificity makes the species conservation-reliant. The strong research-management connection throughout the conservation efforts to recover and sustain this species was only possible because of the collaborative data sharing and commitment to the collection and use of long-term datasets. The banding work and nest studies prior to listing provided important life history information upon which recovery strategies were originally developed. While listed, the banding programs and nest studies from 1984-1992, 1995-2001, 2006-2012, and 2014-2019 provided important evaluations of on-going management on the breeding grounds. The banding programs from 2002-2010 and 2017-2019 elucidated critical habitat relationships on the wintering grounds. More importantly, all banding/nesting datasets were shared and supplemented other local and landscape scale studies to provide management recommendations to improve local habitat conditions and landscape habitat configurations throughout the species’ lifecycle to increase habitat use, warbler reproduction and survivorship, and ultimately, population size. These data parameterized population dynamics models allowing sensitivity analyses to illuminate additional research needs, such as improved understanding of movement rates within and between core and periphery sites. Leading up to listing and post-delisting, the long-term datasets informed new research on the future impacts of climate change and alternative management scenarios, and on the necessity of cowbird control as a conservation strategy. The future of the species will continue to rely on the strong research-management connection coordinated by the Kirtland’s Warbler Conservation Team and supported by shared, long-term datasets, as the species faces stochasticity in the ever-changing economic, social and ecological landscape.|
|A National Multi-Scale Assessment of Genetic Degradation Risk for Forest Tree Species|
|Genetic diversity is essential for forest tree species because it provides a basis for adaptation and resilience to environmental stress and change. The loss of the option value conveyed by genetic variation could be particularly detrimental to the future survival of tree species in the face of numerous severe stresses. The fundamental importance of genetic variation is recognized by its incorporation in the Montréal Process, which the USDA Forest Service uses as a forest sustainability assessment framework for the United States. One Montréal Process indicator for the conservation of biological diversity is the “Number and Geographic Distribution of Forest-Associated Species at Risk of Losing Genetic Variation and Locally Adapted Genotypes.” This indicator has been difficult to address in a systematic fashion, however. We leverage two broad-scale datasets to assess this indicator: (1) species occurrence data from the nationally systematic Forest Inventory and Analysis (FIA) plot network and (2) climatically and edaphically defined provisional seed zones, which encompass geographic areas with similar geology, climate, vegetation, soils and hydrology. Specifically, we intersect the FIA data with the provisional seed zones, which are used as proxies for among-population adaptive variation under the assumption that adaptive genetic variation within species is associated with the kind of environmental conditions that define the seed zones. We then determine, for each species, the ratio of mature trees to saplings within each seed zone as an indicator of insufficient regeneration that could lead to the loss of genetic variation. The results offer insights into which species and which areas of the country may be experiencing degradation of genetic diversity. Such erosion of genetic variation makes species less able to adapt to environmental change, increases the risk of extinction, and lowers the overall resilience of forest ecosystems.|
|Analyzing Multiple Long-Term Data Sets for Monitoring Forest Biodiversity in the National Report on Sustainable Forests|
|Mark D. Nelson; Kurt H. Riitters; Carolyn Sieg; Michael S. Knowles; Guy Robertson|
|The Montréal Process (MP) provides a standard international framework for assessing a set of criteria and indicators (C&I) of the sustainability of temperate and boreal forest ecosystems and their ecological, social, and economic components in twelve countries, including the United States. The first of seven criteria addresses Conservation of Biological Diversity, which is organized into nine indicators within three sub-criteria: ecosystem diversity, species diversity, and genetic diversity, consistent with the definition from the Convention on Biological Diversity. Indicators of United States forest-associated biodiversity are compiled by the U.S. Forest Service (USFS) and presented in the National Report on Sustainable Forests. Biodiversity status and trends are analyzed using several long-term datasets. Ecosystem diversity indices are based on USFS Forest Inventory and Analysis (FIA) database and USGS National Land Cover Database. Tree diversity is assessed from FIA, the USDA PLANTS database, and Biota of North American Program (BONAP). Diversity of vertebrates and selected invertebrates is based on NatureServe At-Risk Species database, USGS Breeding Bird Survey (BBS), and other tabular and geospatial datasets. Total area of United States forest land use is increasing over the past decade, with increasing area in large diameter size classes and decreasing area in medium and small diameter size classes, but net loss of interior forest cover. Forest-associated taxa showed a general increase in the proportion of possibly extinct or at-risk species. Declines in number of forest bird species varied regionally. Ten percent of forest-associated species no longer fully occupy their former range; substantially higher rates of range shrinkage occur for infraspecies than for species. We report additional results from the forthcoming 2020 report.|
|Leveraging long-term changes in species composition to understand shifts in patch preference in desert rodents|
|Ellen K. Bledsoe; S. K. Morgan Ernest|
|Long-term monitoring of ecological systems is imperative for understanding how biodiversity changes through time, and long-term data sets with experimental components can be especially useful. Long-term monitoring offers us the ability to leverage natural experiments by capturing rare, stochastic ecological events (e.g., environmental disturbances, species colonization and extinction). The Portal Project, an experimental site in the Chihuahuan desert in southeastern Arizona, was initially established in 1977 to investigate competition between granivorous ants and rodents and for 40+ years we have collected monthly data on the abundance and diversity of the rodent community. The site also conducts a long-term experiment where kangaroo rats (Dipodomys sp.) have been excluded, creating small landscape of patches with and without a major competitor. In the mid-1990s, a novel competitor, Chaetodipus baileyi, colonized and we leverage this event to demonstrate how patch preference in a ubiquitous, congeneric species (C. penicillatus) shifted in response to changes in species composition and altered competitive networks at the site. This unique opportunity to better understand biodiversity dynamics and species interactions in mammals was only possible because a long-term experiment was in place to capture the dynamics in response to a rare, natural colonization event.|
|Challenges (and solutions) for long term biodiversity data collection, analysis, and archiving.|
|Long-term monitoring and research studies on biodiversity have many challenges. These include changes in field data collection planning, data collection methods, statistical data analysis, spatial data analysis, GIS mapping, and data archiving that occur over time. With continuous updates and upgrades in technology, scientific advancements in field methods, analysis techniques, software, and access to field sites, it is important to integrate these changes while the study is ongoing. New data collection tools exist such as laser rangefinders that supplement clinometer and GPS units with more satellites for increased accuracy of plots locations (some of these tools may not have been available when the study began). GIS and statistical software, including open source software, have continuous upgrades. Field personnel change and need regular trainings. All of these require awareness to advance the project over time to keep it relevant and usable for future scientists. From the start in data collection to the end of data archiving, solutions include personal flexibility and consistent documentation at all stages, metadata creation and practices, and personal and professional investment in the project and the benefits for biodiversity conservation.|
Organizers: Mark Nelson, St. Paul, MN; Jenny Rechel, Riverside, CA; Annika Keeley, Berkeley, CA; Greg Smith, Kent, OH; Angela Larsen, Clemson, SC; Matthew Ihnken, Lansing, MI; Alicia Amerson, San Diego, CA
Supported by: Spatial Ecology and Telemetry Working Group, Biometrics Working Group, Invasive Species Working Group