The Seedlot Selection Tool (SST) is a web-based mapping application designed to help natural resource managers match seedlots with planting sites based on climatic information. The SST can be used to map current climates or future climates based on selected climate change scenarios. It is tailored for matching seedlots and planting sites, but can be used by anyone interested in mapping climates defined by temperature and water availability. The SST is most valuable as a planning and educational tool because of the uncertainty associated with climate interpolation models and climate change projections. The SST allows the user to control many input parameters, and can be customized for the management practices, climate change assumptions, and risk tolerance of the user.
The Seedlot Selection Tool is a collaboration between the US Forest Service, Oregon State University, and the Conservation Biology Institute. Initial conceptualization and development was done by Glenn Howe at Oregon State University College of Forestry and Brad St.Clair at the USFS Pacific Northwest Research Station, with considerable input from Ron Beloin while he was working at Oregon State University. The Conservation Biology Institute was brought onboard to bring the project to fruition through their expertise in web site design and programming for spatial applications. Personnel at the Conservation Biology Institute include Dominique Bachelet (project co-PI), Nikolas Stevenson-Molnar (tool developer), and Brendan Ward (project manager).
Initial funding for the Seedlot Selection Tool came from the US Forest Service Washington Office. Subsequent funding came from the USFS Pacific Northwest Research Station, Oregon State University, and the USDA Northwest Climate Hub.
Dr. Glenn Howe – Co-Principal Investigator
Associate Professor, Department of Forest Ecosystems and Society
Oregon State University, Corvallis, Oregon, USA
Dr. Brad St.Clair – Co-Principal Investigator
Research Geneticist, Pacific Northwest Research Station
USDA Forest Service, Corvallis, Oregon, USA
Dr. Dominique Bachelet – Co-Principal Investigator
Senior Climate Change Scientist
Conservation Biology Institute, Corvallis, Oregon, USA
Nikolas Stevenson-Molnar – Tool Developer
Conservation Biology Institute, Corvallis, Oregon, USA
Brendan Ward – Project Manager
Conservation Biologist/GIS Analyst/Software Engineer
Conservation Biology Institute, Corvallis, Oregon, USA
Populations of forest trees and other native plants are genetically different from each other, and adapted to different climatic conditions. Therefore, natural resource managers must match the climatic adaptability of their plant materials to the climatic conditions of their planting sites. Generally, local populations are optimally adapted to their local climates, or nearly so. Thus, local seed sources are usually recommended for reforestation and restoration. Typically, this has been accomplished using geographically defined zones (e.g., seed zones or breeding zones) or seed transfer rules that specify a geographic or climatic distance beyond which populations should not be moved. However, these recommendations assume that climates are stable over the long-term—an assumption that is unlikely given projected climate change. Because populations are genetically adapted to their local climates, the health and productivity of native or newly established ecosystems will likely decline as climates change. Climate models are now available that can be used to define zones based on climate rather than geography, or calculate climatically based seed transfer limits. Once climatic transfer limits have been defined, natural resource managers can explore options for responding to climate change through assisted migration.
To match seedlots and planting sites, it is first necessary to choose appropriate climate variables. This information can come from genecology studies, which are used to understand how seed source climates influences adaptive trait variation among populations via natural selection. In general, genetic studies indicate that temperate plants are adapted to temperature—especially cold temperatures during the winter, warm temperatures during the summer, and moisture related variables such as precipitation and heat:moisture index.
Once important climate variables are selected, it is necessary to decide on transfer limits—that is, how far can we move a population climatically before performance becomes unacceptable. This information can be obtained by measuring growth and survival in long-term field tests (e.g., provenance and reciprocal-transplant studies) that move populations across large climatic gradients. However, because well-designed studies are rare for many plant species, we often must rely on generalizations from other species and practical experience. For example, climatic variation in seed zones and breeding zones can be used to set transfer limits. Many of these zones have been used for decades—solving earlier problems with maladaptation resulting from excessive seed transfer. No matter what method is used, transfer limits should adjusted to reflect the management practices and risk tolerance of the user. Agencies or organizations that are able to adjust management practices quickly (e.g., because of short rotations), or are more willing to accept risk, may choose a wider transfer limit than those that are risk-averse.
The Seedlot Selection Tool (SST) is a web-based mapping application designed to help natural resource managers match seedlots with planting sites based on climatic information. For example, given a planting site, the SST identifies geographic areas that have similar climates. Thus, the resulting mapped areas show where one could collect well-adapted native seed for planting. Alternatively, when these areas coincide with breeding zones, the mapped areas show where to obtain well-adapted materials from breeding programs. Given a seedlot location, the SST can be used to identify geographic areas with similar climates—that is, candidate planting areas where the seedlot is expected to grow well. In each case, the SST defines the center of the climatic space to be mapped, and then maps all areas that fall within a specified climatic distance (climatic transfer limit) based on climate variables selected by the user. Within the mapped area, the degree of climatic similarity is shown using different colors. Areas that fall outside of the transfer limit are not mapped. To run the SST, the user must specific two climatic regimes—the climate to which the seedlots are optimally adapted, and the climate of the planting site. Typically, the user would choose one of the historical climates as the seedlot climate, and a current or future projected climate as the planting site climate. By choosing a future climate, the SST can be used to examine how assisted migration might be used to respond to climate change.
The Seedlot Selection Tool involves the following steps: (1) select objective, (2) select location, (3) select climate scenarios, (4) select transfer limit method, (5) select climate variables, and (6) map your results. By allowing the user to select the climate change scenarios, transfer limits, and climate variables, the results can be adjusted to reflect the management practices, available knowledge of adaptation, and risk tolerance of the user.
Step 1 – Select objective. The SST was designed to help natural resource managers (1) find seedlots for specific planting sites or (2) find planting sites for specific seedlots. Because these two objectives must be approached differently, the first step is to select one of these two objectives.
Step 2 – Select location. The location of interest can be selected using coordinates or by clicking on the map. However, the location has two different meanings, depending on whether you are searching for seedlots or planting sites. The location of a planting site is straightforward—it is the geographic location of the site you intend to plant. The location of a seedlot may refer to two different things. If seed are collected form a specific stand, then the user can specify the exact location of the seedlot using coordinates or by clicking on the map. In other cases, the seedlot may represent seed collected from a larger zone. In this case, the user can enter the location of the zone's climatic center, and then use either the Custom or Zone transfer limit method in Step 4. If the zone's climatic center is unknown, the user can use the Zone transfer limit method to get its location. In this case, the user would enter any location within the desired zone in Step 2, and then choose the 'climatic center' option in Step 4.
Step 3 – Select climate scenarios. The first step is to identify which climate the seedlots are adapted to, which is typically assumed to be the climate having the greatest influence on the seedlot's parents. Two 30-year normals are available: 1961-1990 and 1981-2010. The 1981-2010 normals represent the 'current' climate. For the SST, we obtained climate data using ClimateNA v5.30 and a USGS DEM data at a resolution of 15-arc-seconds (∼450 m). More information on ClimateNA can be found by clicking on the 'ClimateNA' tab under the 'More Information drop-down menu.
The next step is to choose when you want the planted trees to be optimally adapted to their planting site. Typical choices are the current climate (e.g., 1981-2010), or if you want to account for climate change, some future time period. The future time periods available are: 2011-2040, 2041-2070, and 2071-2100.
If a future time period is used, the final step is to select a representative concentration pathway (RCP), which is associated with different levels of atmospheric greenhouse gases and climate change. The two options are RCP4.5 and RCP8.5. According to IPCC AR5, the RCP4.5 “stabilization” scenario has a projected increase in mean annual temperature of 1.8°C by 2100 (range = 1.1-2.6°C), whereas the RCP8.5 “business as usual” scenario has a projected increase of 3.7°C by 2100 (range = 2.6-4.8°C). The projected climates used by the SST are ClimateNA ensemble projections (averages) across 15 CMIP5 models.
Step 4 – Select transfer limit method. Transfer limits can be set by the user using the Custom method or the Zone method. The Custom method might be preferred when there is good existing information on the effects of seed transfer. As described in the Background, this might come from nursery or field-based genetic tests, or from the operational experience gained from long-term planting programs. If appropriate climatic transfer limits are not known a priori, the user can use the Zone approach, which estimates transfer limits from existing seed zones, breeding zones, or other zones that geographically define acceptable transfer distances. Because managers have been using some zones for decades to guide seed transfer, the climatic transfer distances associated with the zones have been empirically tested.9
Select your climatic center (Zone method). If you have chosen to find planting sites for a seedlot, the first step in the Zone method is to specify the climatic center of your mapped output. If you have a seedlot from a specific known location, you would typically use that location as the center of your mapped output. If you have a seedlot that represents an entire seed zone, you will probably want to use the climatic center of that zone.
Select your species of interest (Zone method). The second step in the Zone method is to select your species of interest, which will determine which zones to display as available options. The generic option returns seed zones that are not species specific. Once selected, only seed zones that are either generic or specific for the selected species will show up in the drop-down menu.
Select your zone of interest (Zone method). The third step in the Zone method is to select one of the available zones. Only seed zones that correspond to your selected species and location (Step 2) will be shown. If no zones are available (e.g., the location is outside the region or species range), then the SST will indicate that there are no zones at this location.
Transfer limits. If you use the Custom method, you will be asked to enter transfer limits for each climate variable in Step 5. If you use the Zone method, the SST obtains the transfer limit (TL) for each climate variable using the selected zone: TL = (xmax-xmin)/2, where xmax and xmin are the maximum and minimum climate values for the zone. Because some unusual zones return outlier TLs, we also provide the average TL for all zones in the selected zone set (e.g., species/author combination). This and other information will show up in a pop-up window if you hover over the climate variable in the climate variable table.
Step 5 – Select climate variables. The user can choose among 16 temperature and precipitation related variables available from ClimateNA. As described in the Background, these variables were chosen based on a wide variety of plant genecological studies. For a location to be mapped (i.e., have a climatic match > 0), it must fall within the TL for each selected climate variable (see Step 6). Thus, the more climate variables that are used, the smaller the mapped areas will be. Although, the points that are excluded are those that have extreme values for multiple climate variables, the use of many climate variables will probably result in overly conservative climate matches. Thus, we caution users from selecting too many climate variables, particularly variables that are unrelated to adaptation. It is also best to avoid selecting variables that are very highly correlated with one another.
Step 6 – Map results. The SST uses the Custom or Zone-based transfer limit (TL) for calculating the climatic match. First, the gridded data for each climate variable are re-scaled: y = |x – xmid|/TL, where xmid is the midpoint value, or climatic center. Then, the multivariate climatic distance (d) from the climatic center to each grid point is calculated as the Euclidean distance for n climate variables: dn = (y12 + y22 + ∙∙∙+ yn2)0.5. Finally, the climate match (m) is calculated as m = ‒(d-1)*100. Values of m < 0 are not mapped, whereas values between 0 and 100 are mapped using a color scale ranging from light to dark orange.
ClimateWNA: generating high-resolution climate data for climate change studies and applications in Western North America
ClimateWNA is an application written by Dr Tongli Wang that extracts and downscales 1961-1990 monthly climate normal data from a moderate spatial resolution (4 x 4 km) to scale-free point locations, and calculates monthly, seasonal and annual climate variables for specific locations based on latitude, longitude and elevation. The downscaling is achieved through a combination of bilinear interpolation and dynamic local elevational adjustment. ClimateWNA uses the scale-free data as baseline to downscale historical and future climate variables for individual years and periods between 1901 and 2100.
The monthly baseline data for 1961-1990 normals were compiled from the following sources and unified at 4 x 4 km spatial resolution:
Historical monthly data were obtained from Climate Research Unit (CRU) (Harris et al 2014). The data version is CRU TS 3.23. The spatial resolution is 0.5 x 0.5° and covers the period of 1901-2014. Anomalies were calculated for each year and period relative to the 1961-1990 normals.
Future climate data
The climate data for future periods, including 2020s (2010-2039), 2050s (2040-69) and 2080s (2070-2100), were from General Circulation Models (GCMs) of the Coupled Model Intercomparison Project (CMIP5) included in the IPCC Fifth Assessment Report (IPCC 2014). Fifteen GCMs were selected for two greenhouse gas emission scenarios (RCP 4.5 and RCP 8.5). When multiple ensembles are available for each GCM, an average was taken over the available (up to five) ensembles. Ensembles among the 15 GCMs are also available.
1) Annual variables:
Directly calculated annual variables:
Derived annual variables:
Wang, T., Hamann, A. Spittlehouse, D.L. and Murdock, T.Q. 2012. ClimateWNA – High-resolution spatial climate data for western North America. Journal of Applied Meteorology and Climatology 51:16-29.
Wang, T., Hamann, A., Spittlehouse, D., and Aitken, S. N. 2006. Development of scale-free climate data for western Canada for use in resource management. International Journal of Climatology, 26(3):383-397.
Daly. C., M. Halbleib, J. I. Smith, W. P. Gibson, M. K. Doggett, G. H. Taylor, and J. Curtis, 2008. Physiographically sensitive mapping of temperature and precipitation across the conterminous United States. Int. J. Climatol., 28, 2031–2064.
Harris, I., Jones, P.D., Osborn, T.J. and Lister, D.H. (2014), Updated high-resolution grids of monthly climatic observations - the CRU TS3.10 Dataset. International Journal of Climatology, 34. pp. 623-642.
Issues are used to track to-dos, bugs, feature requests, and more. As issues are created, they'll appear here in a searchable and filterable list. To get started, you should create an issue.