Changes in climate drive recent monarch butterfly dynamics
Erin R. Zylstra, Leslie Ries, Naresh Neupane, Sarah P. Saunders, M. Isabel Ramirez, Eduardo Rendon-Salinas, Karen S. Oberhauser, Matthew T. Farr, and Elise F. Zipkin
Nature Ecology & Evolution
Code/Data DOI: 
Abstract
Declines in the abundance and diversity of insects pose a substantial threat to terrestrial ecosystems worldwide. Yet, identifying the causes of these declines has proved difficult, even for well-studied species like monarch butterflies, whose eastern North American population has decreased markedly over the last three decades. Three hypotheses have been proposed to explain the changes observed in the eastern monarch population: loss of milkweed host plants from increased herbicide use, mortality during autumn migration and/or early-winter resettlement, and changes in breeding-season climate. Here, we use a hierarchical modeling approach, combining data from >18,000 systematic surveys to evaluate support for each of these hypotheses over a 25-year period. Between 2004–2018, breeding-season weather was nearly seven times more important than other factors in explaining variation in summer population size, which was positively associated with the size of the subsequent overwintering population. Although data limitations prevent definitive evaluation of the factors governing population size between 1994–2003 (the period of the steepest monarch decline coinciding with a widespread increase in herbicide use), breeding-season weather was similarly identified as an important driver of monarch population size. If observed changes in spring and summer climate continue, portions of the current breeding range may become inhospitable for monarchs. Our results highlight the increasingly important contribution of a changing climate to insect declines.
R code for 2004-2018 models
- FullModel_2004-2018.R: R code used to model population dynamics of monarch butterflies (Danaus plexippus) in eastern North America between 2004 and 2018. This is the model used for inferences that includes all seasonal covariates.
- HierarchicalPartitioning_Summer_2004-2018.R: R code used to run all models needed to evaluate the relative importance of covariates in the summer submodel. Includes code to run hierarchical partitioning analysis.
- HierarchicalPartitioning_EarlyWinter_2004-2018.R: R code used to run all models needed to evaluate the relative importance of covariates in the winter submodel. Includes code to run hierarchical partitioning analysis.
- ReducedModel_2004-2018.R: R code used to model population dynamics of monarch butterflies in eastern North America between 2004 and 2018. This is a reduced model, similar to that used for 1994-2003 data. Results were used to verify that estimates from a reduced model were consistent with those from a full annual-cycle model using data from the same period.
Stan model files for 2004-2018
- FullModel.stan: Stan model file, used in FullModel_2004-2018.R.
- FullModel_HierPart_Summer.stan: Stan model file, used in HierarchicalPartitioning_Summer_2004-2018.R.
- FullModel_HierPart_Winter_0FE_NoSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- FullModel_HierPart_Winter_0FE_InclSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- FullModel_HierPart_Winter_1FE_NoSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- FullModel_HierPart_Winter_1FE_InclSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- FullModel_HierPart_Winter_2FE_NoSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- FullModel_HierPart_Winter_2FE_InclSummer.stan: Stan model file, used in HierarchicalPartitioning_EarlyWinter_2004-2018.R.
- ReducedModel_2004-2018.stan: Stan model file, used in ReducedModel_2004-2018.R.
R code for 1994-2003 models
- ReducedModel_1994-2003.R: R code used to model population dynamics of monarch butterflies in eastern North America between 1994 and 2003. This is the model used for inferences that includes all seasonal covariates.
- HierarchicalPartitioning_1994-2003.R: R code used to run all models needed to evaluate the relative importance of covariates in the summer submodel. Includes code to run hierarchical partitioning analysis.
Stan model files for 1994-2003
- ReducedModel_1994-2003.stan: Stan model file, used in ReducedModel_1994-2003.R.
- ReducedModel_HierPart.stan: Stan model file, used in HierarchicalPartitioning_1994-2003.R.
Data files for 2004-2018 (full annual-cycle model)
All monarch data from the overwintering grounds and covariate data are publicly available. Monarch data from the summer breeding grounds are proprietary and are therefore not publicly available (though we provide descriptions of those files here).
- Monarchs_winter.csv: Data on overwintering monarch aggregations, 2004-2018.
- supercolony.ind: a unique index identifying each supercolony (1:13)
- supercolony: name of each supercolony
- reserve: indicates whether a supercolony is inside (1) or outside (0) of the butterfly reserve
- yr: year
- area: area occupied (ha) by monarch butterflies
- forest.dense.p: percent of surrounding area with dense forest cover
- Covariates_County.txt: Covariate values associated with each county on the summer breeding range, in the U.S. and Canada.
- county.ind: a unique index identifying each county (1:545)
- state.county: combined state-county FIPS code
- state: FIPS code for each state
- county: FIPS code for each county in a state
- state.name: 2-letter code for each state
- county.name: name of each county
- area.land.sqmi: land area of each county (sq. mi)
- Xcentroid: longitude of county centroid
- Ycentroid: latitude of county centroid
- avgGDD: average of annual GDD values accumulated between 3 May and 15 Aug (weeks 10-24), 2004-2018
- avgPCP: average of annual, cumulative precipitation (mm) between Apr and Aug, 2004-2018
- perc.open: percent of each county that is unforested
- perc.crop: percent of each county associated with agricultural crops
- surveyed: indicates whether one or more monarch surveys were conducted in the county between 2004 and 2018 or not (1 or 0, respectively)
- Covariates_Year.csv: Annual covariate values, 2004-2018.
- yr: year
- feb: total area occupied (ha) by monarchs on the overwintering grounds in late February
- spGDD.east: annual GDD values accumulated between 22 Mar and 2 May (weeks 4-9) in eastern Texas.
- spPCP.east: cumulative precipitation (mm) between Feb and Apr in eastern Texas.
- NDVIR1: mean NDVI for the first (northern) half of autumn migration.
- Covariates_CountyYear.csv: Annual covariate values associated with each county on the summer breeding range, in the U.S. and Canada, 2004-2018.
- county.ind: a unique index identifying each county (1:545)
- state.county: combined state-county FIPS code
- yr: year
- glyphosate: estimated proportion of corn and soy crops in each county sprayed with glyphosate
- diffPCP: difference between annual precipitation (cumulative, Apr-Aug) and average precipitation (average of annual values, 2004-2018) in each county
- Covariates_CountyWeek.csv: Weekly covariate values associated with each county on the summer breeding range, in the U.S. and Canada, 2004-2018.
- county.ind: a unique index identifying each county (1:545)
- state.county: combined state-county FIPS code
- yr: year
- wk: week (16-24)
- diffGDD: difference between GDD and average GDD (2004-2018) for that week and county
- Monarchs_summer.csv: Data from monarch surveys on the summer breeding grounds, 2004-2018 (not publicly available).
- program: name of monitoring program
- state.county: combined state-county FIPS code
- lat: latitude of survey location
- long: longitude of survey location
- yr: year
- wk: week (16-24)
- monarch: total number of adult monarchs observed during survey
- duration: number of person/party hours spent surveying
- site.ind: a unique index identifying each survey location
- county.ind: a unique index identifying each county
- perc.open: percent of area immediately surrounding each survey location that is unforested
Data files for reduced annual-cycle models (1994-2003 or 2004-2018)
- Monarchs_winter_1994-2018.csv: Data on overwintering monarch aggregations, 1994-2018.
- yr: year
- dec: total area occupied (ha) by monarchs on the overwintering grounds in late December
- dense: percent of surrounding area with dense forest cover, averaged among supercolonies
- Covariates_County_1994-2018.txt: Covariate values associated with each county on the summer breeding range in the U.S.
- county.ind: a unique index identifying each county (1:502)
- state.county: combined state-county FIPS code
- state: FIPS code for each state
- county: FIPS code for each county in a state
- state.name: 2-letter code for each state
- county.name: name of each county
- area.land.sqmi: land area of each county (sq. mi)
- Xcentroid: longitude of county centroid
- Ycentroid: latitude of county centroid
- avgGDD: average of annual GDD values accumulated between 3 May and 15 Aug (weeks 10-24), 1994-2018
- avgGDD.9403: average of annual GDD values accumulated between 3 May and 15 Aug (weeks 10-24), 1994-2003
- avgGDD.0418: average of annual GDD values accumulated between 3 May and 15 Aug (weeks 10-24), 2004-2018
- avgPCP: average of annual, cumulative precipitation (mm) between Apr and Aug, 1994-2018
- avgPCP.9403: average of annual, cumulative precipitation (mm) between Apr and Aug, 1994-2003
- avgPCP.0418: average of annual, cumulative precipitation (mm) between Apr and Aug, 2004-2018
- perc.open: percent of each county that is unforested
- perc.crop: percent of each county associated with agricultural crops
- surveyed: indicates whether one or more monarch surveys were conducted in the county between 2004 and 2018 or not (1 or 0, respectively)
- Covariates_Year_1994-2018.csv: Annual covariate values, 1994-2018. File structure is the same as that of Covariates_Year.csv except that February estimates of area occupied on the overwintering grounds (feb) and NDVI values in autumn (NDVIR1) are not included because they are unavailable during this time period.
- Covariates_CountyYear_1994-2018.csv: Annual covariate values associated with each county on the summer breeding range, in the U.S., 1994-2018.
- county.ind: a unique index identifying each county (1:545)
- state.county: combined state-county FIPS code
- yr: year
- glyphosate: estimated proportion of corn and soy crops in each county sprayed with glyphosate
- diffPCP: difference between annual precipitation (cumulative, Apr-Aug) and average precipitation (average of annual values, 1994-2018) in each county
- diffPCP.9403: difference between annual precipitation (cumulative, Apr-Aug) and average precipitation (average of annual values, 1994-2003) in each county
- diffPCP.0418: difference between annual precipitation (cumulative, Apr-Aug) and average precipitation (average of annual values, 2004-2018) in each county
- Covariates_CountyWeek_1994-2018.csv: Weekly covariate values associated with each county on the summer breeding range, in the U.S., 1994-2018.
- county.ind: a unique index identifying each county (1:545)
- state.county: combined state-county FIPS code
- yr: year
- wk: week (16-24)
- diffGDD: difference between GDD and average GDD (1994-2018) for that week and county
- diffGDD.9403: difference between GDD and average GDD (1994-2003) for that week and county
- diffGDD.0418: difference between GDD and average GDD (2004-2018) for that week and county
- Monarchs_summer_1994-2018.csv: Data from monarch surveys on the summer breeding grounds, 1994-2018 (not publicly available). File structure is the same as that of Monarchs_summer.csv.