From Caprio's lilacs to the USA National Phenology Network



Continental-scale monitoring is vital for understanding and adapting to temporal changes in seasonal climate and associated phenological responses. The success of monitoring programs will depend on recruiting, retaining, and managing members of the public to routinely collect phenological observations according to standardized protocols. Here, we trace the development of infrastructure for phenological monitoring in the US, culminating in the USA National Phenology Network, a program that engages scientists and volunteers.

Changes in the timing of seasonal events – such as flowering, migrations, and breeding – can serve as a “globally coherent fingerprint of climate-change impacts” on organisms (Parmesan 2007). Climate-induced changes in phenology have been linked to shifts in the timing of human allergy seasons and cultural festivals, increases in wildfire activity and pest outbreaks, shifts in species distributions, declines in the abundance of native species, the spread of invasive species, changes in agricultural yield, and changes in carbon cycling in natural ecological systems. Phenological data can also provide critical information needed for understanding important issues, such as agricultural and wild plant species not meeting their requirements for exposure to cold temperatures in winter, timing mismatches for interacting species, and agricultural adaptation. Even in the US, phenological data are limited, and existing long-term datasets tend to be species- or site-specific. Therefore, although climate is a known critical driver of phenological variation of organisms across scales from individuals to landscapes, we are generally unable to answer ecologically and societally important questions, such as: (1) how do phenological variations in time and space affect the abundance, movement, distribution, genetics, and interactions of organisms?; and (2) can we forecast phenological responses to climate variability and change across populations and interacting species in both managed and unmanaged ecosystems?

The future of phenological research, monitoring, and understanding will depend on a coordinated effort to organize and collect phenological and related information (eg climatological and hydrological data) across a variety of spatial and temporal scales. Phenological monitoring activities in the US will be most successful if they are integrated with other ecological science and monitoring networks, remote-sensing products, emerging sensor technologies and data management capabilities, and formal and informal educational opportunities; active participation by members of the general public will also be necessary. The success of these monitoring efforts will depend on how well they inform science, resource management, and policy, as well as the degree to which they empower the public in formulating and facilitating adaptive responses to a changing climate (Figure 1).

Figure 1.

Phenology is one of the most sensitive biological responses to climate change, is a critical part of nearly all aspects of ecosystem function, and is relatively easy to observe, requiring little specialized monitoring equipment.

Organized phenology monitoring in the US: a brief history

The first spatially extensive phenological observation networks in the US, focused on lilacs and honeysuckles, were initiated by the US Department of Agriculture (USDA) in the late 1950s and early 1960s to characterize seasonal weather patterns and improve predictions of crop growth and development (Schwartz 1994). The volunteer observers included a few thousand cooperative weather service observers, scientists and technicians at agricultural stations, and garden club members, who provided data on leafing and flowering phenology via the US mail. Encouraged by the success of a program established in 1956 by Joseph Caprio (Montana State University) in the western US, similar projects (later merged into a single eastern network) were initiated in the central and northeastern states in 1961 and 1965, respectively. Observations in the western network continued until 1994 (Caprio retired in 1993), while the eastern and central networks were terminated in 1986 after losing funding. Additional details about historical regional phenology networks are presented in WebPanel 1.

Rebirth of the lilac network: an incipient national network

The western US lilac network was reactivated in 1997 by Dan Cayan and Mike Dettinger of the Scripps Institution of Oceanography and the US Geological Survey, respectively, to complement their studies on changes in timing of snowmelt discharge (Cayan et al. 2001). Similarly, Mark Schwartz (first author of this paper) reactivated the eastern US lilac network in 1988 for climatological research. Figure 2a provides a synopsis of the spatial and temporal distribution of the historical lilac phenology datasets across the continental US between 1956 and 2008.

Figure 2.

(a) Years of observations for USA Lilac Phenology Stations, 1956–2008. (b) The multi-taxa phenology monitoring program, Nature's Notebook, has ~4000 registrants at ~5000 sites tracking ~16 000 organisms across the nation (as of October 2011). These sites (shown only for the continental US because of space limitations, and excluding lilac observation stations from panel a) include those maintained by members of the public, local schools and clubs, and neighborhood associations, as well as US National Parks and National Fish and Wildlife Refuges, the National Ecological Observatory Network and the Long Term Ecological Research Network research sites, and nature preserves.

In 2002, Schwartz started planning to revitalize and broaden the lilac networks into a national framework, while also extending phenological observations to other native and non-native plant species. Following a workshop to discuss how the incipient National Ecological Observatory Network (NEON) might contribute to monitoring and understanding the ecological impacts of climate change (AIBS 2004), Julio Betancourt (coauthor of this paper), David Breshears of the University of Arizona, and others echoed the idea of continental-scale phenological monitoring. Independent of NEON, in 2005, Schwartz, Betancourt, and colleagues began to develop a national network of phenological observation stations, and existing lilac and honeysuckle observation stations were reorganized to form a prototype Plant Phenology Program for that network.

As proof of concept for a more extensive network, the observations of lilac and honeysuckle phenology made by thousands of volunteers since the 1950s have contributed many useful insights about spring onset variations at regional to continental scales (Schwartz 1994; Cayan et al. 2001; McCabe et al. 2011). Three examples of inferences made from lilac phenology at the continental scale are provided in Figure 3. There is a high correlation between observed “first leaf” dates and “first bloom” dates (Figure 3a), which is stronger in colder climates and weaker in warmer ones. For species that behave similarly to lilacs, “leaf out” may be a suitable predictor of flowering and conceivably could be used to forecast associated phenomena such as pollen production and allergen loads. Figure 3b shows the North American continent-wide annual average variation (in days) of modeled first leaf dates from weather stations that report daily maximum and minimum temperatures across the continent. These Spring Indices are calibrated and validated with observations of first leaf and first bloom dates for lilac and honeysuckle, and show an abrupt advance in the timing of spring onset in the mid-1980s. The regional risk of an early or late spring, based on positive or negative phases of El Niño Southern Oscillation and the Pacific Decadal Oscillation in the prior October through December, is also estimated (Figure 3c; McCabe et al. 2011).

Figure 3.

(a) Relationship of lilac first leaf to first bloom observations in North America. Axis units represent day of the year, with January 1st = 1. (b) Modeled first leaf departures averaged across North America. (c) Risk of early modeled first leaf date for (i) positive October through December (OND) NINO3.4 (a measure of sea surface temperature variations in the central equatorial Pacific, bounded by 120°W–170°W) and positive OND Pacific Decadal Oscillation (PDO) conditions, and (ii) negative OND NINO3.4 and negative OND PDO conditions from 1900 through 2008 (from McCabe et al. 2011).

Development of the contemporary network

The contemporary USA National Phenology Network (USA-NPN; was established in 2007 with support from the US National Science Foundation, the US Geological Survey, and several other agencies and organizations. The USA-NPN is a consortium of individuals and organizations that collect, share, and use phenological data, models, and related information. Its mission is to serve science and society by promoting a broad understanding of plant and animal phenology and its relationship with environmental change. Volunteer observers collect data on hundreds of species, including the common and cloned lilac species observed by network precursors, across the nation (Figure 2b). In turn, USA-NPN makes phenology data, models, and related information freely available to scientists, resource managers, and the public to aid in decision making and adapting to changing climates and environments. Additional details about USA-NPN are presented in WebPanel 2.

Numerous other phenology observation programs in the US operate either independently of or in cooperation with the USA-NPN. Although many were developed within the past 10 years, some of these programs have been operational for several decades; for example, the North American Bird Phenology Program engaged thousands of volunteers to track migratory bird phenology across the continent between 1880 and 1970. The geographic scope of contemporary programs ranges from international (eg eBird, Global Learning and Observations to Benefit the Environment), national (eg FrogWatch USA, Journey North, Project BudBurst), and regional (eg Eastern Pennsylvania Phenology Project, Signs of the Seasons), to state or local (eg Ohio State University Phenology Garden Network, Penn Phen) initiatives. Moreover, contemporary projects have a variety of programmatic missions, including science, education, and/or public engagement. The particular focus may vary, with some programs concentrating on tracking phenology of specific species or taxa, whereas others collect many different types of observations. Coordination and collaboration among the diversity of programs across the nation, while retaining individual programmatic identity and stakeholder value, will be both a challenge and an opportunity in the coming decades.

Sustaining a national phenology observing system

The reasons for the success or failure of environmental monitoring networks are not always apparent (but see Lovett et al. 2007); often, such networks are heroic efforts that wax and wane with their champions, whether individuals or organizations. For example, the Western States Phenological Network lasted from 1956 to 1994, benefiting from Caprio's position as an agricultural meteorologist at Montana State University and generous support from both the USDA and the National Weather Service. But when Caprio retired in 1993, the program faltered because of a lack of personal and institutional interest. A decade or so later, other individuals emerged to develop USA-NPN, an effort much broader in motivation, scope, and interest than its predecessors.

What steps can be taken to ensure the success of networks such as USA-NPN? Support from the US Federal Government will be critical because federal agencies typically outlast individuals, are responsible for long-term and nationwide planning, and are not subject to the lapses in funding faced by academic institutions and non-governmental organizations. Monitoring efforts usually have to strike a balance in scope and avoid trying to satisfy too many diverse objectives. Breadth of scope is usually an advantage early on but can become a liability as a network matures and base funding and participation stabilize. Communication with stakeholders and their changing needs is critical for maintaining the relevance of networks over time.

Two measures of network success will be the number, distribution, and retention of loyal and capable observers and the strategic value of observations across the continent. To focus coverage on national and regional information needs, network expansion will need to evolve beyond attraction of volunteers through mass marketing and toward directed recruitment and management of practiced observers who will record specific phenological events for a focused list of species in targeted locations. For example, the current distribution of USA-NPN observation stations, which grew arbitrarily, is correlated with human population densities (Figure 2b). To achieve a more homogeneous distribution of stations nationwide and to avoid heat islands and other urban effects, the USA-NPN may have to refocus recruitment and retention activities in rural communities. Also, phenological information becomes optimal when recorded near sites where other environmental variables are monitored, including weather, radiation, biogeochemical fluxes, hydrology (especially soil moisture), and plant and animal demographies. Although uniformity in monitoring is a desired goal at the national level, monitoring efforts should also vary depending on long-term regional trends and projections. For example, over the past 50 years, spring has advanced several days in the Northeast, Upper Midwest, and western US but has been delayed in the Southeast, the result of a so-called “warming hole” that arises from internal Pacific Decadal Oscillation variability and could persist under climate change (Meehl et al. 2012). These and other continental-scale differences in historical and predicted climate should be addressed by regional and national campaigns to monitor and study phenology.

Finally, to ensure long-term success, the USA-NPN must consider and balance the needs of: (1) its observers – both volunteer and professional – typically working at local scales; (2) land/resource managers and private enterprise normally operating at landscape to subregional scales; and (3) researchers and policy makers interested in the science and management of global change at all scales. Principles for collaboration among network participants include mutually beneficial activities, shared vision about science and education, realistic demands on the capacities of partners, feedback to improve collaboration, and transparent data and information sharing policies.


We dedicate this contribution to the memory of the late Joseph M Caprio. The USA-NPN gratefully acknowledges sponsoring organizations: US Geological Survey, University of Arizona, University of Wisconsin–Milwaukee, The Wildlife Society, US National Park Service, National Oceanic and Atmospheric Administration, National Aeronautics and Space Administration, National Science Foundation (IOS-0639794), Oak Ridge National Laboratory, and US Fish and Wildlife Service. Staff from the USA-NPN National Coordinating Office collaboratively developed the description of the contemporary network in the body of the text and in WebPanel 2.