Natural enemies research cluster discussion

Importance of natural enemies to monarch dynamics and spatiotemporal patterns of distribution

1) Main questions - data resources/gaps

Data resources/gaps:

Shawna Stephen's thesis data from CA 2002-2003 (Dennis will contact)

Chip's graduate student (mid-1990s) from U of Kansas?

Master parasite database updates (Karen, Becky)

Reports of other disease? viral? bacterial?

MLMP--'dead from other causes'?

Future sampling protocols for rearing larvae to test for parasitoids (i.e., Hawai'i)

Reports of disease and parasitoids in other danaids--UGA (Sonia, Becky), check w/Myron (Becky), Leong?

Paul Cherubini and David Marriott may have data on Oe along the CA site gradient - look at Western monarch discussion group (had scanned hand sheets-Dennis knows how to look)

Questions:

1) What is causing the decrease in OE in western N Am populations? Is this pattern related to observed change in population size? -

1a) Other potential mechanisms--i.e., milkweed distributions/availability, species composition, effects of drought, effects of monarch releases?

-insufficient data to examine effects of releases

1b) How could changing growing seasons impact OE? via milkweed abundance and accumulation of spores on plants?

1c) What are the density-dependent effects on horizontal transmission and overall OE prevalence? Based on size of population

Paper: see notes below

Data:

Oe: 1997, 2002 (Shawna), 2003-2009 (mostly from Jaap).  2003 might be breeding.  Still need data from Shawna/Dennis for 2002.

Monarch abundance: 1997-2009 annual estimates of wintering adult abundance from Thanksgiving counts (Dennis)

1990-2010 wintering adult abundance data also available from CA Fish and Game.  May need to first check correlation between 2 data sets, and then combine data sets for longer series if appropriate.

Working hypotheses:

A) The decline in Oe infection levels is related to decreasing monarch populations (transmission-limited)

B) The decline in Oe infection levels is related to environmental drivers

-Spore viability could be reduced because of drought conditions (see Sonia's earlier experimental work?) as a shortened growing season may lead to lower transmission.

-Infected monarchs may be more susceptible to drought conditions, resulting in differential mortality and potential density-dependent infection dynamics

     -see Dennis and K. Leong's work survival data with changing environmental conditions 1996-2010

     -Art Shapiro's data on declining butterfly trends in general in CA as a result of landuse change (urban development) in CA over the past 20   years

-->Direct effects on monarchs from decreased rainfall and decreased irrigation could reduce Oe levels through

1) Increased relative mortality of infected individuals (drought-intolerant)

2) Decreased spore viability (see UGA summer 2010 experiments, and plant tapings)

-Organize/analyze old humidity expt (Sonia)

-Organize/analyze plant tapings for spore #s (Becky)

-Acquire CA F&G data (Dennis)

-Organize CA Thanksgiving counts and Shauna's 2002 data (Dennis, Becky)

-Organize Oe data (Becky)

-Time series in Oe (Becky, Sonia)

-Time series in monarch abundance

-Correlation between Oe/monarch abundance w/env variables as covariates (Becky, Sonia?)

2) What is causing the increase in OE in eastern N Am populations? 

2a) What effect of increasing distribution of winter breeding densities along the Gulf coast?

-Andy and Elizabeth reported that JN occurrence data from the Psyche paper (2010) is available as a time series from 2002-2010 for overwintering sites in the south.  There is some limited abundance data associated with this series as well

-MLMP density data as a possibility too (Karen) 1997-present  Need to show increases between mid-late 1990s and present.

2b) In years with cold winters followed by increasing levels of OE?

-Is there a winter thermal minina along the SE coast where monarchs could breed year round?  Use winter temperature indices to create an index of suitability and correlate with interannual variation in Oe

    -If there are increasing levels of wintering monarch populations, are Oe levels increasing?  If so, spring migrants make encounter infected residents as they return north

    -If milkweed is patchily distributed and more aggregated, populations may be more infected with Oe due to increased transmission...data unavailable?  see Zalucki and Lammer 2010

Data:

-Oe trends from early 1960s to present

-Monarch abundance trends in terms of annual index (MLMP? NABA? WWF?)

-Organize Oe data (Becky)

-Decide on abundance index and acquire series/get permission of use if necessary

-Acquire temperature data for winter thermal layer (see Leslie?)

-Correlate annual OE prevalence with average wintering count estimates – relative index or scale

Paper: see notes below

Papers:

For publishing questions #1 and 2, there are 3 options:

*1) 1 large paper integrating long-term Oe trends of both eastern and western populations in addition to evaluating difference in infection between migratory and non-migratory populations (Costa Rica, Australia, Hawaii, Puerto Rico, Miami).  Authors: Sonia, Dennis, Jaap, Karen, and Becky.  Author order unknown.

Advantages: The contrast of comparing E (where infection increases) vs. W (where infection decreases, Oe could possibly go extinct)

2) 2 papers: 1 evaluating N. Am trends in Oe and the other looking at mig vs. non-mig populations

3) 3 separate papers

*seems to be preferred option at the moment (Oct 2010)

3) How does the series of natural enemies during monarch development interact? Oe, early predators, tachinid flies, pupal parasitoids.

Effects of being a vertically-transmitted parasite in an infected host - can you prime the host immune system to go after a tachinid larva or a pupal parasitoid?

What drives low levels of Oe in CR and Puerto Rico? What are the relationships between parasitoids and OE prevalence?  In Costa Rica (an outlier in the global OE trends), are the observed lower patterns of OE infection related to high levels of parasitoid infection (i.e., larvae are not surviving to adulthood to test for OE infection) - at this point, data collection issue. 

Hypotheses: Are both responding to monarch densities?  Plant densities and patch dynamics (mw is very patchy in CR)?  Self-limiting disease that leads to purges?  Is the infective time of a spore different in the tropics?

Need ongoing monarch density measurements from CR - MLMP through UGA program there?  Karen talk to Nate about setting up a research program, Becky will tell him where milkweed is, and he could map distribution?

Are oe infected larvae more likely to succumb to tachinids?  Need experimental work on this.  Karen will contact Michelle Solensky about methods.  Jaap student Eleanor is maybe working on this, but we're not sure what she's finding.  We could ask her about her specific questions. 

Contact other lepidopterists working in Costa Rica to see what they know about tachinid hosts - not many people rear larvae.  Author of caterpillar book - David Wagner (U of CT) might have information - Karen will contact him david.wagner@uconn.edu.  Also contact editor of tachinid times.  Faculty member who works on Leps and tachinids - Jeff Boettner.  Other people in Lep Soc.

Could one of Nate's students look at this while he's in Costa Rica.  Scott Connelly at CR station - may have a student who could do a project. Mike Maudsley also.

need historical data on Oe abundance in CR

4) What are the effects of patch dynamics on natural enemy dynamics (patch isolation, proximity, dispersion)?  For example, what are the impacts of a single infected female monarch on an isolated patch? or effects of a tachinid potentially infecting an entire isolated patch?  Individual-based modeling efforts?  Check Hawai'i (Becky will take the lead)

4a) How do both of these infections respond to spatiotemporal patterns of monarch abundance?  (based on count data or MLMP)

Data gap: parasitoid infection levels in other butterfly species

Data gap: TX, FL participants sampling of MH/MLMP in resident (non-migratory) populations?

Data gap: PR efforts for MLMP through MH volunteers--newsletter (Becky) and instruction form for both programs (Becky, Karen).  For example, MLMP add contact info for MH OE and for MH to include parasitoid sampling efforts.  Change datasheets.

patch model - look at plant distribution and dispersion to get data for model

Becky, Sonia (submitting proposal in Jan 2011 - maybe include parasitoids), Nate, Karen

5) What are the effects of different predators (specifically fire ants and wasps) on monarch dynamics? How do different enemies interact?  How do enemies at one life stage affect the next life stage? Review/survey paper - Karen lead

5A) How does this vary for specialist and generalist natural enemies?

Questions:
-How to think about marginal mortality? mi = 1 – (1 – q)^qi/q where qi = mortality from agent i and q = all mortality during stage
-How to think of additive or compensatory mortality?
-Specialists will be more tightly linked to host, generalists cause "spillover" mortality, and may be agents of apparent competition, although there are many possible outcomes of shared predation
-Is mortality from pathogens increasing?

-Can predators keep populations healthy (see C. Packer's work)

-Preferential predation? Potential disease culling

Modelling:

Life stage model - conceptual diagram of how enemies would affect each other (see predator list.xls updated on wiki)

--Something that infects during one stage will be negatively affected by something that kills or had sublethal effects during that or subsequent stages
--Something that kills during one stage may be affected (pos and neg?) by something that infects during an earlier stage (if the earlier enemy affects palatibility, nutritional value, etc)
--Modeling could be done from a host perspective, or an enemies perspective
--Density-dependence during any stages

--Additive and compensatory mortality – mortality early in life has a higher chance of being compensatory than later in life, which is more harmful
--Consequences of host rarity and generalist specialist issue
--Rob Dunn and Kevin Gross (NCSU) working model on specialists and generalists within a single host

Data:

Generalists: wasps, lacewing larvae (expts from Matt?), ladybug larvae, spiders, ants, bugs, mantids, bacteria, viruses, data from Alma?

-Data on wasp predation in monarchs – Karen is working on one, Linda Rayor working on this in larvae; Karen lost whole experiments to wasp predation; Wasps eat pupae, adults
-Adult predators:  mantids and birds
-Spiders, ants, hemipterans that attack the early stages of larvae
-Generalist pathogens – nematode parasites, mainly viruses, fungi, and bacteria that can attack monarchs but are not specialists; kill larvae, Js and pupae; what are these diseases and are they on the rise?

Specialists: Oe, tachinids 1997-2010, MLMP?

-OE killing – reduced eclosion, shortened adult lifespan – this is for doses that are realistic ecologically; weak evidence for increased development time increase; on average, what is the number of days this is increased?  [Sonia look this up – based on what we have from past lab experiments]

-Tachinids – lethal at the J-pupal stage

-Literature search on interactions between enemies – for example, the predator-disease papers and ‘keeping the herds healthy and alert’  - predators can drive down the prevalence of disease and actually increase the abundance of prey (Sonia)

-Organize tachinid data (Karen)

-Collate other generalist data

Paper:

Opinion piece - sketch out conceptual framework and research agenda.  TREE?  Feature monarch as case study, look at other systems (Karen lead)

5B) What are the effects of fire ant predation on monarch survival?

- If we find an effect of fire ants then we could do a follow up analysis of landscape variable effects (esp those that index fire ant control by humans) and climate effects (esp rainfall) within the fire-ant infested areas

Paper: ‘Fire ants create a dead zone for monarch recruitment in North America’--as a Note in Biology Letters? Sonia, Karen, Becky, Dennis

Data:

Ants: Use fire ant first sightings at county level – 2004, have GIS layer from Matt Fitzpatrick
-Can we get any more recent data or any other measures of fire ant abundance (mound density)? Becky will email Rob about this
-Maybe fire ant abundance data exists at a more local scale for a subset of areas, to do a more focused analysis

Monarch survival: data from MLMP
-Need to decide what is the best index of survival – survival from egg to 5th instar, survival to earlier instars, or Weibull index if aggregate at a larger spatial scale.  Or could calculate the Weibull for each site. 
-For Weibull index, need to use time or degree days – so use the same value of Zalucki’s degree days across all the sites

Criteria for excluding monarch survival data:

–number of milkweeds > 10

-number of counts > 4 and < 30 (or look at where they live to see how many weeks they should realistically be looking at...exception of Texas?)
-If number of 5ths per egg is > 1 then set to 1, and > 2 then exclude the data totally
-Could weight everything by the number of eggs (so if see few eggs, don’t count as heavily in analysis)

-Analysis would have to control for year ants were first sited at a county

Analyses/Task list:

-MLMP data: add Lat/Long, other instars, sampling effort, site info (Karen)

-Check on additional ant data (Becky)

-Give MLMP spatial context in ArcGIS and overlay MLMP and ant data (Becky, Nate)

-Identify landscape variables that correlate with monarch survival: (latitude, size of site, milkweed species present, management on land, type of site, surrounding area, other milkweed sites nearby, is milkweed planted?)--Karen, James

-Check for an interaction effect with fire ants and residential areas, open habitats, etc.

- If we find an effect of fire ants then we could do a follow up analysis of landscape variable effects (esp those that index fire ant control by humans) and climate effects (esp rainfall) within the fire-ant infested areas 

6) Parasitoids - Karen will take lead

Are large scale ups and downs related to overall population size? MLMP survival, count data, size of winter population, NABA, BMNs

Driven by densities of other hosts?  Contact tachinid researchers - how many hosts are there, can we find data on abundance of other hosts,

Contact Michelle Solensky about what she's doing.

weather: winter survival - how do they overwinter

Data resources :

MLMP? Alma's work (see Karen)

Lacewing functional response (Matt Kaiser)

Data gaps:

earwigs? 

Calvert's work on ants in The Monarch Butterfly book

2) Result highlights - discussions
3) Generality to other migratory species

Migratory escape

Migratory culling

4) Planned ms (preliminary authorship)

Review paper on monarch predators?

Resistance to natural enemies?