As ectotherms, temperature plays an important role in the development of insects. For insect pests, all else being equal, faster development and more generations are expected be associated with greater damage. Prior research has estimated the degree-day requirement for GWSS (the amount of time spent above the temperature required for development from an egg through to adult), which has been used to explore why GWSS is more of a problem in some areas or some years (Pilkington et al 2014). Here, we’ll use such degree-day modeling to consider whether conditions in recent years favored GWSS becoming more of a problem, then check that against trapping data in the region.
Using air temperature data from a Temecula weather station going back more than 40 years, I estimated the number of GWSS generations that could be supported in a given year. The figure above shows the mean value for all years prior to 2012 (“historical”) relative to each year separately since 2012. Historically, temperatures in Temecula were sufficient to support 2 GWSS generations most years, with an “extra” 3rd generation about once every three years on average. However, since 2012 degree-day accumulation has been sufficient to support a 3rd generation in 6 of 8 years. In other words, “extra” generations are predicted to have been common in recent years at a higher frequency that was historically the case.
In addition to calculating the potential number of generations, I also estimated how quickly the degree-day requirement for GWSS was achieved each year. The assumption is that early completion of a generation means more time for the pest to be active and cause damage – especially for GWSS, since early season infections are more likely to lead to chronic infection and Pierce’s disease in grapevines. The figure below shows how long it took to reach the degree-days required to get through the 2nd generation. Historically, it took about 250 days (early September). A couple of the more recent years took a little longer than that (above the top dotted line), a couple were about the same (between the dotted lines), but half of the recent years took significantly less time. In 2013, for example, it’s estimated to have taken more than 2 months less time.
Both of the analyses shown above are troubling in that they indicate that temperatures in recent years may be favoring more rapid development of GWSS, perhaps in a way that contributes to more damage. But what does the areawide GWSS trapping data say?
First, I compared yearly trap catch to the yearly estimated number of generations, with the expectation being they should be positively associated with each other (more generations = more GWSS on traps). The figure below shows, in red/brown, the mean number of GWSS adults caught per trap over the year, for each year since 2012. From this, it is apparent that 2013 and 2015 were the lowest-catch years and 2017 was by far the highest. Indeed, since the values are square-root transformed, 2017 is quite a bit more extreme than it looks in the figure. In blue are the predicted number of generations from the degree-day modeling, which show the highest values in 2013 and 2016 and lowest values in 2017 and 2018. In other words, # of GWSS caught and # of generations don’t seem to be related to each other in a meaningful way.
Finally, I compared predictions of how quickly the degree-day requirement was met with patterns in the Temecula trapping data. In the figure below, the green line is the day of the year associated with the peak number of GWSS caught that year (140=early June, 240=late August). 2019 is notably delayed relative to the other years, which mostly had peak catches in early to mid-July. The red line shows the estimated time to reach the degree-day requirement for two GWSS generations (from Fig. 2). This value is more variable, with the highest values in 2012 and 2017, and the lowest value in 2013. Again, these two sets of values that you might expect to be correlated, don’t seem to be.
In summary, the degree-day modeling suggests that conditions in recent years were sufficient to support a high frequency of “extra” GWSS generations and more rapid onset of those generations in Temecula, which is troubling if they continue to hold in the years ahead or become more pronounced (2013 was estimated to be more than half-way to 4 generations). And yet, the trapping data doesn’t indicate that those extra generations necessarily translate to greater numbers of GWSS. This discrepancy may stem from ongoing IPM for GWSS in citrus and vineyards. Alternatively, it may suggest there are more important forces driving GWSS resurgence or year-to-year variation in activity than temperature-dependent development.
Removal of infected grapevines is often considered an important element of Pierce’s disease management to limit their potential to serve as a source of the pathogen, Xylella fastisiosa, for insect vectors. The Fall is an ideal time to scout vineyards for PD given that symptoms are most obvious this time of year.
The following handout describes what symptoms to look for when scouting for PD, along with other biotic and abiotic diseases that might be confused with it: