Time to Restore

Pollinator restoration presents many challenges, from selecting which species to plant to provide nectar during critical periods, to anticipating how these plant species will respond to changes in climate.

A better understanding of flowering and seed timing for critical nectar plants, and the links between this activity and climate, can inform more resilient restoration plantings. 

We are a team of collaborators from the Bosque Ecosystem Monitoring Program, the Tribal Alliance for Pollinators, the Gulf Coast Phenology Trail, and the USA National Phenology Network, supported by a grant from the South Central Climate Adaptation Science Center. Our project, Time to Restore: Connecting People, Plants, and Pollinators, aims to deliver guidance to those working on pollinator restoration. This Story Map describes our work over the past two years to (1) better understand the needs of those in the pollinator restoration community in the South Central Region, (2) collect data on flowering and seed timing of nectar plants, and (3) deliver this information in a format that is useful to those in the restoration community.

Want to be part of the project and give your input? Email erin@usanpn.org to be added to the Time to Restore Quarterly Call list.

We began our project with a needs assessment of those working on pollinator restoration in three states in the South Central Region - New Mexico, Oklahoma, and Louisiana.

In the Fall of 2021 we hosted a regional kickoff workshop and three state-level workshops. Our participants included members of federal and state agencies, Tribes and Pueblos, Master Gardeners, conservation organizations, native plant groups, universities, and more.

The goals of our workshops included:

• Understand the decision-making process for restoration and how different organizations and groups deal with variable scales, budgets, and other challenges
• Understand climate changes occurring in each state and the projected changes for coming decades
• Understand how changes are impacting plant flowering and seed timing and what we can learn from data collection platforms that record these seasonal events
• Discuss best practices for Indigenous data sovereignty, considering perspectives of individual Tribes and Pueblos
• Provide a place for those working on pollinator restoration and conservation in the South Central region to connect
• Get input from participants on priority species, monitoring guidelines, and locations to target for data collection efforts across the state beginning in early 2022

We learned the following information about the existing decision-making process for restoration.

Three key needs emerged from stakeholder workshops:

We set out to tackle the need for additional data collection on the timing of flowering and seeding in plants that our participants care about. We first explored the available scientific literature within the field of phenology.

Phenology is the study of the timing of life cycle events of plants and animals and their relationship to the climate and other factors like daylength.

We explored the phenology data available for nectar plants in the region, as well as published research that explored the timing of flowering and seeds and how changes in climate impact the timing. Multiple factors can influence the timing of phenology events in plants such as flowering, including warmth, freeze events, winter chill, rainfall, and daylength. Generally, researchers have documented shifts to earlier flowering in many plants (United Nations Environment Programme, Frontiers 2022, Wittington et al 2015).

While little research has been done in the region, authors of studies in other locations have found a large shift, such as common milkweed in the Midwest and Northeast shifting 7 days earlier with each °F warming (Howard 2018). Other research found a different pattern between spring and fall-flowering species in the Southeastern Coastal Plain. Spring species flowered 3-4 days earlier per °F of warming while fall species flowered slightly earlier with warmer spring temperatures and later with warmer summer temperatures at a rate of 2 days per °F (Pearson 2019). Under experimental warming, prairie plants flowered 2-10 days earlier under the higher level of warming of 5.4 °F (Wittington et al 2015).

We asked workshop participants about what kind of information would be most useful in their restoration planning process. They indicated a need for all metrics of flowering and seeding.

We surveyed our stakeholder community about how they would like data presented in a visual format. Stakeholders expressed interest in both a community-style calendar that displayed information about flowering and fruiting timing, as well as a single species calendar with more detailed information about the peak in timing.

All of our workshop participants indicated an interest in collecting data on flowering and seed timing of nectar plants. We also asked about the platform that participants were most likely to use for data collection. The most popular platform was iNaturalist, though participants in all three states also voted for Nature's Notebook. Multiple participants in New Mexico elected to maintain their own data privately. We promoted both iNaturalist and Nature's Notebook as options for data collection.

Due to the lack of data in the region, we needed to determine a short-list of priority species to focus data collection efforts. During workshops, participants came up with a list of species important to restoration in their state. We then had participants vote on which species were of greatest interest, and the following species received the most votes:

1. wild bergamot (Monarda fistulosa)
2. cardinal flower (Lobelia cardinalis)
3. green antelopehorn (Asclepias viridis)
4. common sunflower (Helianthus annuus)
5. eastern purple coneflower (Echinacea purpurea)
6. buttonbush (Cephalanthus occidentalis)
7. showy milkweed (Asclepias speciosa)
8. tall blazing star (Liatris aspera)

To support data collection, we held a Train the Trainers workshop and distributed training and outreach materials to recruit data collection participants. Additional materials can be found on the Time to Restore landing page:  https://usanpn.org/TimeToRestore 

During the period of January 2022 - June 2023, we collected observations on four of our eight focal species at 13 sites. In our data analysis, we included any observations that had been contributed to Nature's Notebook prior to 2022 and also expanded the list of species to include eight additional species that were received a high number of votes during our workshops.

The map below displays the sites collecting data via Nature's Notebook on these 16 species, by genus. The size of the circle represents the number of years of data collection at that site.

The table below shows the number of site years (count of "open flowers" or "ripe fruits" onset events recorded for an individual site or year) by species.

One of the main needs identified in our workshops was information on how flowering and seed timing will shift under future climate conditions. The South Central Climate Adaptation Science Center provides information about how temperatures, precipitation, and other aspects of climate are expected to shift over the coming century. We used the climate projections for mid-century (2036-2065) as the most relevant for pollinator restoration in the coming decades. We also acquired projections at the seasonal level directly from SC CASC staff to use in our analysis to predict how phenology is likely to shift under future climate conditions.

In this recorded presentation, Alyssa Rosemartin explains projected changes in climate in the region and the potential impact on plants and their pollinators. Alyssa presented this information as part of our Fall 2021 Stakeholder Workshops.

We created individual species calendars and multiple species calendars for each state. Both types of calendars display data in a "heatmap" format. The calendars plot observations contributed to Nature's Notebook and display the weekly proportion of records where an observer reported "yes" or "no" for either open flowers or ripe seeds. Darker colors indicate higher proportions of "yes" records for that week. Gray colors indicate that all of the records submitted in that week were "no"; no color indicates that no data were collected that week.

Single species calendars include a heatmap as well as two line graphs. The top line graph displays all of the data contributed to iNaturalist from 2011 to June 2023 for that species. The second line graph displays data contributed to Nature's Notebook on the number of open flowers (a value calculated by combining the number of flowers and percent of open flowers).

For each of the species in our calendars, we explored relationships between climate variables and flowering and seed timing. For flowering and seed timing variables, we included metrics that stakeholders indicated were most relevant - start of open flowers, peak in open flowers, duration of open flowers, start of ripe seeds, peak in ripe seeds, and duration of ripe seeds. We used several methods for ensuring data quality, such as only including data where there was a "no" record prior to the first "yes" record for flowers or fruits within 14 days.

Based on existing research, and knowing that we wanted to connect our findings with climate projection data, we selected average maximum and minimum temperatures by season as well as seasonal precipitation totals as our predictor variables. We created simple linear models to identify relevant climate variables, then created linear mixed effect models. We set data collection site as a random effect, as we expect plants at the same site to behave similarly.

Due to the lack of phenology data in the region, we expanded our analysis to include all phenology data on our species of interest collected via Nature's Notebook across the Conterminous United States in our models. We included a variable for latitude to ensure that we accounted for any variation across latitude in the relationship between climate cue and flowering or seed metric. The map below shows the location of data included in the climate models for species that had sufficient data.

Our linear models identified climate cues (such as Fall Precipitation and Spring Minimum Temperature) with a strong relationship with life cycle stages (such as Open Flowers Onset and Fruit Peak Duration). We multiplied the effect sizes from the models (e.g. 1.8 day earlier per degree F warming) by the projected changes to climate (e.g. 3 degree increase in average low temperature by mid century) to determine an estimated projection of shift in timing for each species and life cycle stage.

The table below displays the full list of projected shifts in flowering and fruiting from our models.

See the full set of climate models results and projections.

To summarize all of this information - published research, climate projections, flowering and seed calendars, and climate cues - we created three Info Sheets, one for each state.

Phase II launches this fall, Time to Restore: Developing people-driven tools for climate-smart restoration.

The main focus of Phase II will be:

• Additional data collection on priority species to fill in gaps in data and allow us to create calendars and models of shifts in flowering and seed timing at a more local level
• Addition of species, particularly bosque species of importance to Pueblos in New Mexico
• Addition of Texas to the project, including a stakeholder workshop and Texas coordinator to establish long-term data collection sites
• Bring in herbarium data to increase temporal depth of data
• Updated phenology calendars, info sheets, StoryMap
• Pilot of seed notification service in Oklahoma
• Pilot of climate-smart planting palette tool in one ecoregion

We invite you to be part of Time to Restore for this next phase of the project. Email erin@usanpn.org to be added to the Time to Restore Quarterly Call list.

Project Team Erin Posthumus, USA National Phenology Network, PI Alyssa Rosemartin, USA National Phenology Network, Co-PI Gail Bishop, Gulf Coast Phenology Trail Jane Breckinridge, Tribal Alliance for Pollinators Robert Croll, Great Lakes Indian Fish and Wildlife Commission Kim Eichhorst, Bosque Ecosystem Monitoring Program Gina Lloyd, Louisiana Coordinator for Time to Restore April Taylor, South Central Climate Adaptation Science Center Sue Wilder, Gulf Coast Phenology Trail Hayley Limes, student, USA National Phenology Network Travis Matlock, student, USA National Phenology Network

References: 

Dixon K.W., A.M. Wootten, M.J. Nath, J. Lanzante,  D.J. Adams-Smith, C.E. Whitlock, C.F. Gaitán, R.A. McPherson, 2020: South Central Climate Projections Evaluation Project (C-PrEP), South Central Climate Adaptation Science Center, Norman, Oklahoma, USA.  doi.org/10.21429/12gk-dh47  Howard, A.F. 2018. Asclepias Syriaca (Common Milkweed) flowering date shift in response to climate change. Sci Rep 8,17802.  doi.org/10.1038/s41598-018-36152-2  Pearson, K.D. 2019. Spring- and fall-flowering species show diverging phenological responses to climate in the Southeast USA. Int J Biometeorol 63, 481–492. doi.org/10.1007/s00484-019-01679-0 Whittington, H. R., D. Tilman, P. D. Wragg, and J. S. Powers. 2015. Phenological responses of prairie plants vary among species and year in a three-year experimental warming study. Ecosphere 6(10):208. dx.doi.org/10.1890/ES15-00070.1

United Nations Environment Programme 2022. Frontiers 2022: Noise, Blazes and Mismatches – Emerging Issues of Environmental Concern. Nairobi.