An introduction to DDRP

Invasive pests present a significant threat to agricultural production in the United States, yet decision support tools that can accurately predict where and when to expect pests have not yet been fully developed and utilized. Our spatial modeling platform known as DDRP (Degree-Days, Risk, and Phenological event mapping) was designed to provide regularly updated forecasts of the potential distribution (risk of establishment) and timing of seasonal activities (phenology) of pests (Barker et al. 2020,Barker et al. 2023). An overview of the DDRP modeling process is below (Fig. 1).

Fig. 1. DDRP modeling overview

Species models

Currently we are using DDRP to produce regularly updated (every 2-3 days) forecasts for 18 high-priority invasive insect species, available at USPest.org (Fig. 2). The platform has also been adapted to predict the phenology and voltinism (number of generations per year) of three biological control insects that have photoperiod-cued diapause (Grevstad et al. 2022), also available at USPest.org. We are developing moisture processing features in DDRP to improve the modeling of moisture-sensitive organisms such as plant pathogens and weeds.

Fig. 2. DDRP models developed to date.

Inputs and outputs

DDRP uses a process-based (mechanistic) approach to model temperature-dependent development, phenology, and climate suitability of target species (Fig. 1). The platform requires gridded daily minimum and maximum temperature data, and information on the temperature requirements for development and survival of a species. We typically run DDRP using current and forecast climate data for the conterminous U.S. to provide real-time decision support for a species; however, the platform accepts data for any time frame or region, such as data for past years or for other countries. Model products include maps of the predicted potential distribution (climate-based risk of establishment), number of generations, and dates of phenological events. The potential distribution is represented by areas where cold and heat stress accumulations have not exceeded the stress limits of a species.

Program features

Some of the major features of DDRP currently include:

1. Degree-day parameters including durations and lower and upper developmental thresholds for four separate life stages (these are the egg, the larva or nymph, the pupa or pre-oviposition, and the adult), plus a separately parameterized overwintering stage.
2. The ability to spread the population using cohorts. Typically seven cohorts are specified but any number can be used. While cohorts offer the ability to spread the population in a Gaussian or other distribution, there is currently no distributed-delay function, meaning that the spread does not increase over multiple generations.
3. Phenological event maps (PEMs, also known as pest event maps), which depict estimated calendar dates of seasonal activities or population events. PEM parameters are specified as degree-days within each of the four (plus overwintering) stages. For example, DDRP can be parameterized to make first egg-hatch PEMs by setting a degree-day value near the completion of the egg stage, or at the beginning of the larval stage. If the former is used, then a second PEM, say for mid-larval development, could be parameterized using a value such as one-half of the degree-day total for larval development.
4. Climatic suitability maps, which show two levels of climatic suitability (moderate and severe stress exclusions). These are intended to indicate risk likelihood of short vs. long-term establishment but could also indicate migration zones, and uncertainties such as in species parameterization, model structure, and in the sources of climate data.

Setup and usage

DDRP is an R script (“DDRP_v3.R”) and must be within the same directory an auxilliary R script that contains program functions (“DDRP_v3_funcs.R”). DDRP has not yet been formatted into an R package because we designed the code to be run from the command line on a Linux server. The program can also be run on a Windows system but parallel processing capabilities will be limited. The user manual (Coop and Barker 2024) “DDRP_user_guide_and_platform_requirements_V5.doc” is the only instruction document that is currently available The instruction manual provides information on program requirements, input data, input options, examples of command line arguments, types of output files, and run times.

Philosophy

Our development of DDRP has strived to achieve a parsimonious balance of both model simplicity and accuracy, with a focus on four philosophies:

1. Simplicity, in that existing data and results for well-studied, major invasive threats can be readily adapted for use
2. Universality, to accommodate a wide range of organisms
3. Robustness, by having the emphasis on use of first-principles that lend to process-driven rather than statistical correlation-driven models
4. Practicality, by focusing on models that can be used for decision support rather than more complex research-only models

Example outputs

The movie below shows DDRP outputs of the emergence of overwintered adults of emerald ash borer over the course of 2021 (Fig. 3). Areas where heat or cold stress has exceeded the stress limits for the species are predicted to be excluded from the potential distribution.

Fig. 3. DDRP forecasts for emerald ash borer for 2021.

Another way to look at this (mostly) same information is with a phenological event map, below (Fig. 4).

Fig. 4. A phenological event map for emerald ash borer for 2021.

Acknowledgements

The development of DDRP was funded by the USDA APHIS PPQ Cooperative Agricultural Pest Survey (CAPS) and Center for Plant Health Science and Technology (CPHST) programs, the USDA National Institute of Food and Agriculture (NIFA), and the Department of Defense Strategic Environmental Research and Development Program (SERDP).

References

Barker, B. S., L. Coop, T. Wepprich, F. Grevstad, and G. Cook. 2020. DDRP: real-time phenology and climatic suitability modeling of invasive insects. PLoS ONE 15:e0244005. https://doi.org/10.1371/journal.pone.0244005.

Barker, B. S., L. Coop, J. J. Duan, and T. R. Petrice. 2023. An integrative phenology and climatic suitability model for emerald ash borer. Frontiers in Insect Science 3:1239173. https://doi.org/10.3389/finsc.2023.1239173

Coop, L., and B. S. Barker. 2024. Computing infrastructure requirements and user guide for hosting DDRP models. Prepared for APHIS PPQ and other collaborators. Available here.

Grevstad, F. G., T. Wepprich, B. S. Barker, L. B. Coop, R. Shaw, and R. S. Bourchier. 2022. Combining photoperiod and thermal responses to predict phenological mismatch for introduced insects. Ecological Applications. 32:e2557. https://doi.org/10.1002/eap.2557