EFFECTS OF HEAT WAVE TIMING ON PLANT, HERBIVORE, DISEASE INTERACTIONS By J oshua S Snook A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Entomology Masters of Science 202 0 ABSTRACT EFFECTS OF HEAT WAVE TIMING ON PLANT, HERBIVORE, DISEASE INTERACTIONS By Joshua S Snook Climate change research has primarily focused on the consequences of increasing average global temperature . However, e xtreme weather events are another consequenc e of climate change that has been harder to examine because of their rar ity and unpredictability . Heat waves are brief events of extreme he at that pose an immedia te threat to food security and agricultural sustainability and are predicted to increase exponentially in frequency over the next fifty years . The effect of heat stress on single organism s are well known, but w e have a poor understanding of how the timing of heat wave events influence plant - insect - disease interactions in agroecosystems . In thi s thesis, I address two primary research questions : 1) H ow does timing of a heat wave impact crop - pest interactions? 2) H ow are multi - species interactions influenced by heat waves? I used environmental chambers in the lab and open - top chambers fitted with infrared heaters in the field to simulate heat wave conditions at early and late development stages of potato ( Solanum tuberosum ) inoculated with an associated herbivore the Colorado potat o beetle ( CPB, Leptinotarsa decemlineata , ( Say ) (Coleoptera: Chrysom elidae) ) and foliar disease early blight caused by ( Alternaria s olani ) . I found that heat waves increased CPB larval mass by 26 - 44% and sped up development time , suggesting there could be an additional generation of herbivor e s per season . Heat waves combin ed with herbivore and disease stress caused 20 - 40% lower yield on potatoes in early reproductive stages . Actions should be taken to protect crops in early reproductive stages from stress through diligent ir rigation practices and more frequent scouting to i nform pest management strategies prior to and during heat waves . iii ACKNOWLEDGMENTS I would like to thank m y advisor s , Zsofia Szendrei and Will Wetzel, f or their mentorship and guidance though the completion of this program. They have provided me with excell ent suggestions on experiment design, implementation, management, and analysis that has broadened my knowle dge and abilities as a scientist. They encouraged me to dig deeply and think critically about this system and the ecological significance of my work. I would also like to thank my committee member Jamie Willbur for her vast knowledge of plant disease resea rch methods and for preparing the inoculum used in the experiments. I would like to thank my wife Andrea Glassmire for her support and encouragement throughout this project . I would also like to thank my lab members and technicians in both the Szendrei and Wetzel labs for their friendship, feedback, and support especially on the long hot days in the field. The following post docs, grad and undergrad st udents , and technicians have all played a major role in this work and I am extremely grateful for their help : Anna Julie, Corrine Johnston, Emily Mall, Cory Snook, Kayleigh Hauri, Elizeth Cinto Me jia , Dan Turner, Moria Robinson, Annie Levardsen, Haley Dole , Wissam Jawad , Jack Rumery, Margie Lund, Patrick Stillson, Logan Appenfeller, Tom Wood, Eli Bloom , Natalie Constancio, Jennifer Savalnitskaya, & Luke Zehr . I would also like to thank t he farm staff at the Kellogg Biological Station, especially farm manage r B rook Wilke and technician Josh Dykstra , for field preparation . Thank you to the Michigan State Entomology department staff for their support during my degree program. I would also like to thank my funding sources : the Michigan Potato Industry Commission , the MSU GREEEN grants program , the W.K. Kellogg Biological Station , and MSU AgBioResearch . Lastly, I would like to thank my friends and family for their love and support during my time here at MSU . iv TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ......................... v i LIST OF FIGURES ................................ ................................ ................................ ..................... vi i CHAPTER 1: E ffects of heat wave timing on plant, herbivore, disease interactions .................... 1 I NTRODUCTION ................................ ................................ ................................ ............... 1 CHAPTER 2 : The effect of heat wave timing on plant - herbivore interaction s .............................. 4 ABSTRACT ................................ ................................ ................................ ......................... 4 INTRODUCTION ................................ ................................ ................................ ............... 5 METHODS ................................ ................................ ................................ .......................... 7 Experimental o rganisms ................................ ................................ ........................... 7 Growth cham ber experiment ................................ ................................ .................... 7 Field experiment ................................ ................................ ................................ ...... 9 Heat wave methods ................................ ................................ .................... 11 Statistical analyses ................................ ................................ ................................ . 1 2 RESULTS ................................ ................................ ................................ .......................... 1 3 Growth chamber experiment ................................ ................................ .................. 1 3 Beetle performance ................................ ................................ .................... 1 3 Potato yield ................................ ................................ ................................ 1 4 Field experiment ................................ ................................ ................................ .... 1 6 Beetle performance ................................ ................................ .................... 1 6 Potato yield ................................ ................................ ................................ 1 9 DISCUSSION ................................ ................................ ................................ .................... 20 A PPENDIX ................................ ................................ ................................ ........................ 2 4 CHAPTER 3 : Consequences of heat waves on plant, herbivore, and disease interactions .......... 2 7 ABSTRACT ................................ ................................ ................................ ....................... 2 7 INTRODUCTION ................................ ................................ ................................ ............. 2 8 METHODS ................................ ................................ ................................ ........................ 2 9 Experimental organisms ................................ ................................ ......................... 2 9 Insect ................................ ................................ ................................ .......... 2 9 Disease ................................ ................................ ................................ ....... 30 Plant ................................ ................................ ................................ ........... 31 Experimental d esign ................................ ................................ ............................... 31 Heat wave methods ................................ ................................ .................... 32 Statistical analyses ................................ ................................ ................................ . 33 RESULTS ................................ ................................ ................................ .......................... 34 Bee tle performance ................................ ................................ ................................ 3 4 Larval survival ................................ ................................ ........................... 3 4 Larval mass ................................ ................................ ................................ 3 5 Larval development ................................ ................................ .................... 3 6 Disease development ................................ ................................ ............................. 3 7 Potato yield ................................ ................................ ................................ ............ 3 9 v DISCUSSION ................................ ................................ ................................ .................... 40 APPENDIX ................................ ................................ ................................ ........................ 45 CHAPTER 4: Conclusions and future directions ................................ ................................ .......... 4 7 APPENDIX ................................ ................................ ................................ ........................ 50 LITERATURE CITED ................................ ................................ ................................ ...... 52 vi LIST OF TABLES Table 2.1 The number of replicates by month of planting, heat wave timing and presence or absence of Colorado potato beetle larvae in a field experiment at the Michigan State University Kellogg Biological Station in 2018. We generated heat wave conditions in field plots using ceramic heaters and open - top chambers each with 10 potato plants (5/month ). . 25 Table 2.2 Mean ± SEM temperatures (°C) for 3 - hour increments during heat of the day and cool of night during field heat wave experiments in Michigan in an experi mental potato field in 2018. Heat waves were created using open top chambers and ceramic heaters. . 25 Table 2.3 Temperature (°C) differences between heat wave and control treatments in open top chambers for heat waves cr eated in an experimental potato field in Michigan in 2018 . 25 Table 3.1 Mean ± SEM temperatures (°C) for field heat wave experiments in Michigan in an experimental potato field in 2019. Heat waves were created using open top chambers and ceramic heaters. Day means are calcula ted during a three - hour interval from (14:00 - 16:00 h), night from (4:00 - 6:00 h) on all five days of the heat wave treatment. Means were calculated from 5 Hobo data loggers in control and 5 data loggers in heat wave tre atments. Also showing Relative humidit y for the same time frames (no standard error only one logger per treatment). . 4 6 Table 3.2 Differences between heat wave and control treatments in open top chambers for heat waves created in an experimental potato fiel d in Michigan in 2019 . .. . 4 6 vii LIST OF FIGURES Figure 2. 1. Survival and mass of Colorado potato beetle larvae feeding on plants exposed to an early heat wave and late heat wave treatments. Blue circles denote larvae feeding on control plants and red triangles represent larvae feed ing on plants exposed to a heat wave treatment. ( A ) Larval survival was lowest in the late heat wave treatment, in which larvae were present on plants during the heat wave. Larval survival was slightly reduced by the early heat wave treatment in which the heat wave occurred on plants one week prior to larval addition. ( B ) Larval mass did not differ between control and heat wave treatments. Points and error bars show means + / - one standard error. Means with diffe rent letters are significantly different from each other (Statistical test p<0.05) .. 1 4 Figure 2. 2. Average tuber mass produced per plant for all heat wave treatment combinations after 70 d. Blue circles denote control plants and red triangles represent plants exposed to a heat wave treatment. The early heat wave had the greatest negative effect on tuber mass. The addition of 3 beetle larvae had no effect on yield compared to plants without larvae and of the same heat wa ve treatment. The addition of 3 beetle larvae to the late heat wave treatment did decrease tuber production compared to control groups. Points and error bars show means + / - one standard error. ........... . . 15 Figure 2 . 3. Average survival of Colorado potato beetle lar vae feeding on different age plants for (A) July and (B) August heat wave treatments in a field at the Kellogg Biological Station. Blue circles denote control plants and red triangles represent pla nts exposed to a heat wave treatment. Larval survival was l ower for those that experienced a July heatwave regardless of plant age. There was no difference in survival of larvae between control and heatwave treatments for the August heat wave, but the larv ae that were feeding on younger 10 - week - old plants and expe rienced a heat wave had a much higher survival rate than those that were feeding on the 12 - week - old plants and experienced a heat wave . . .. . 17 Figure 2. 4. Mean mass of Colorado potato beetle larvae feeding on different age plants exposed to (A) July and (B) August heat wave in a field at the Kellogg Biological Station. Blue circles denote control plants and red triangles represent plants exposed to a heat wave treatment. Larval mass was 30% higher for larvae that fed on 8 - week - old potatoe s and were exposed to a July heat wave (P<0.05) compared to non - heat wave controls. Heat waves significantly influenced larval mass when modeled collectively ( 2 =6.0094, df=6, P= 0.0142 3) 18 Figure 2. 5. Blue circles denote control plants and r ed triangles represent plants exposed to a heat wave treatment. (A) Heat wave treatment significantly reduces marketable tuber mass in 8 - week - old plants that experienced a July heat wave. (B) Marketable tuber mass in the August heat wave was significantly influenced by plant age but heat wave and beetle treatments had no effect. Blue circles denote non heat wave controls and red triangles denote heatwave treatments 20 Figure 3.1 . Mean (+/ - SEM) survival of Colorado potato beetle larvae feeding on co ntrol plants (blue dot), early blight inoculated plants (brown dot), heat wave (HW) plants (red dot), and plants exposed to both early blight and heat wave treatments (yellow dot) . 35 viii Figure 3.2. Mean mass of Colorado potato beetle larvae feedin g on control plants (blue dot), early blight inoculated plants (brown dot), heat wave (HW) treated plants (red dot), and plants exposed to both early blight and heat wave treatments (y ellow dot) in a field at the Kellogg Biological Station. 3 6 Figure 3.3 . Proportion of surviving larvae that achieved 4th instar feeding on control plants (blue dot), early blight inoculated plants (brown dot), heat wave (HW) treated plants (red dot), and plants exposed to both early blight and heat wave treatments (yellow do t). Larvae were placed on plants as first instars and removed after 10 d , with the heat wave treatment occurring from day 3 to day 8. 37 Figure 3.4 . Sum of percent dead potato foliage fo r all seven sampling dates combi ned. The circles represent plants with no CPB treatment and the triangles represent those that received 15 first instar Colorado potato beetles (blue shapes = control and red = heat wave (HW) treatment). ... .. . 38 Figure 3.5 . Average total tuber mass produced per plant for control (blue points) and heat wave (HW) plants (red points) exposed to beetle (CPB), early blight inoculation treatments (Blight), and both beetle and early blight exp osure (Blight+CPB). Circles denote plants without CPB treatments and triangles represent plants exposed 15 first instar Colorado potato beetles . Points and error bars show means + - one standard error from 20 replicates per treatment 40 1 CHAPTER 1: Effects of heat wave timing on plant, herbivore, disease interactions I NTRODUCTION While a gradual increase in global mean temperatures is a widely appreciated aspect of climate change, concomitant increases in climatic variability and extreme weather e v ents pose a much more immediate threat to humans and the resources we depend on ( Bisbis et al., 2018; Meehl and Tebaldi 2004 ; Rahmstorf and Coumou 2011 ; Fischer and Knutti 2015 ; Perkins - Kirkpatrick and Gibson 2017 ; Price 2020 ). Drought and heat waves are predicted to impact 11% and 10% of global crop and grassland with Europe, Africa and the Middle East, and Asia being areas hardest hit (Myers et al., 2017). T echnological advancements in agriculture like large scale irrigation systems have reduced the dama ge of heat waves on agriculture but g rowing enough crops to sustain an increasing human population is a major challenge of the modern world, with food scarcity being a widespread problem (IPCC, 201 4 ). New farming practices and resilient crop varieties are constantly being developed to maintain optimal productivity of the landscape. However, most food crops are produced in specialized monocultures making them especially susceptible to disease and herbivore pest outbreaks, as well as extreme events like f lood ing, drought, and heat waves. Millions of dollars are spent controlling agricultural pests every year yet much of how extreme events like heat waves a ffect pest management is largely unknown (Alyokhin , 200 9 ) . Growth chamber studies exposing crops, ins ects , and pathogens to a temperature gradient identifies their thermal minima, optima, and maxima, but predicti ve models derived from such studies typically fail to incorporate fluctuating temperatures and interspecific interactions (Ferro 2 et al. 1985; Chaeran i & Voorrips, 2006; Logan et al., 1985; Levy & Veilleux, 2007 ). Daily and seasonal fluctuations create an ever - changing environment in which organisms must live and adapt, with some organisms exhibiting behavioral movements to avoid stressful temper atures and intense sunlight, making it difficult to predict the outcome of an extreme event on mobile organisms. Plants also have mechanisms to avoid the negative effects of heat stress by lowering leaf temperatures through transpiration, but periods of lo w preci pitation preceding and accompanying heat waves reduce available soil moisture limiting this protective process (Robinson 2001). The combination of high heat and low precipitation have caused massive yield losses leading to development of highly soph isticat ed irrigation systems to combat heat and drought conditions (Moriondo & Bindi, 2007) . This technology has greatly reduced yield loss of crops like potatoes but this is not a sustainable solution . T he timing of an extreme event could drastically alter the impacts of the event on phytopathogens, insect herbivores, and their host plants , depending on when the event occurs during the ontogeny of each organism and the temporal sequence of their interacti ons (Yang and Rudolf 2010) . A deeper understanding of the timing of heat waves on the interaction between plants and their pests is vital for maintaining sustainable agriculture and resilient natural systems. Th is thesis addressed three primary r esearch qu estions : 1) what are the direct and indirect effects of he at waves on plant insect interactions? ; 2) how does timing of a heat wave alter its impact s on plant pest interactions? ; and 3) how do heat waves alter inset pest, disease pest, host plant interactions? To address the first question , I conducted a growth c hamber experiment to identify the direct and indirect effects of heat waves on plant - insect interactions. With clear direct and indirect effects of heat waves on CPB survival and potato yie ld identified , next I developed a 3 field experiment that exposed fou r different age classes of potato to both heat wave and herbivore pressure to identify the role of heat wave timing on plant insect interactions. I used the specialist herbivore CPB in our experiments. I chose the larval life stages to use in my experiment s because they can be found on plants soon after emergence through till plant senescence , while adult beetle abundance is more variable. Larvae also lack the hardened exoskeleton of adults making them more susceptible to desiccation from heat. Lastly, in t he summer of 2019, I incorporated potato early blight , which is a common foliar disease, into a fully crossed field experiment that identified the impact of heat waves on potato ( Solanum tu berosum ), insect herbivore ( CPB ) , and disease interaction. A lternar ia solani is a fungal pathogen that affects solanaceous plants like potato, tomato, pepper, and eggplant as well as other wild night shade species (Thomma, 2003) . It persists in dead plant tissue from previous nd field margins. Its spores are transferred by wind and rain droplet splash to new host plants. Young solanaceous plants however are considered resistant until near flowering when plant re sources begin to be used for reproduction (Tsedaley, 2014) . I focus ed the experiment around early reproductive life stages of potato tuber initiation and early tuber bulking when the disease begins to infect the lower leaves (Douglas, 1972) . These experiments combined identify key life stages of potato that are especially susceptible to heat waves and herbivores, and gain insight into complex interactions with plant disease that will aide in the preservation of sustainable agriculture in the face of a changing climate. 4 CHAPTER 2 : The effect of heat wave timing on plan t - herbivore interaction s ABSTRACT Brief periods of intense heat and decreased precipitation (heat waves) are becoming more common globally. The immediate negative effects of heat stress on plants and insects are well understood, but we have a poor unders tanding of the potentially lasting e cological effects of heat waves on plant - insect interactions. An understanding of how heat waves alter the dynamics of plant - herbivore interactions in agroecosystems would improve our ability to manage pests in an increa singly variable climate. I examined how heat waves influence the growth and development of potatoes and the associated herbivore CPB. I experimentally imposed heat wave conditions at multiple times during potato ontogeny in the lab in environmental chamber s and in the field using open - top ch ambers and ceramic heating elements. I found that heat waves decreased CPB larval survival on young plants . Moreover, survival was also reduced for larvae that fed on plants that had previously experienced a heat wave, e ven if the larvae themselves did not experience the heat wave. Despite the negative consequences of heat waves for CPB performance, potato plants in early reproductive life stages exposed to both CPB herbivory and heat waves had significantly reduced yield . Our results indicate that even though heat waves may reduce the performance of insect pests, they may also act interactively with the stress of herbivore damage to worsen the effects of pests on plant growth . This suggests that best management pract ices should prioritize reducing CPB populations prior to forecasted heat wave events especially for potatoes in early reproductive stages. 5 INTRODUCTION While climate change biology has mainly focused on gradual change s in mean temperatures ( Bale 2002; Dyer and Letourneau 2013 ; Leisner et al. 2018; Brown et al. 2004; Harrington et al. 1999 ; Parmesan and Yohe 2003 ) , there has been less attention on the impact s of extreme weather events, including heat waves, which have be en increasing in frequency, duration an d intensity in recent years (Meehl and Tebaldi 2004; Rahmstorf and Coumou 2011; Fischer and Knutti 2015; Perkins - Kirkpatrick and Gibson 2017). While there is no widely accepted single definition of a heat wave , for purposes of this work we describe them as extreme d of stressfully high temperatures . Heat waves have been estimated to have occurred less than 4 times per decade per region of the U.S. in 1951 1990 but they are predicted to become more common in many parts of the w orld (Robinson 2001). Heat waves are important not only for their impacts on urban environments and natural ecosystems but also because they have potential impacts on food security, especially since major food crop s in the world are produced in specialize d monocultures. These agroecosystems are fine - tuned to produce maximal yields and often lack resilience to rebound after extreme weather events. Because they result in extreme conditions that may exceed some physiol ogical tolerances, they can disrupt ecolo gical processes (Smith 2011) in ways that may alter pest control. Indeed, extreme weather has already had severe negative impacts on agriculture (Bisbis et al. 2018), including the reduction of national cereal produ ction by 9.1% by heat waves during 1964 2 007 (Lesk et al. 2016). These reductions can be the result of direct and indirect effects on interacting plants and their herbivores in crop fields. Predicting the net effect of an extreme event on pest control requ ires examination of how the individual ef fects of an event combine and 6 interact across trophic levels. To do this we need studies that explore both the direct, physiological effects of events on individual organisms and the ecological effects that occur wh en organisms interact in field settings. A major challenge to understanding how heat waves affect crop - pest interactions is that the sensitivity of organisms likely changes through different life stages. Heat waves occurring at different life stages could have different consequences for the plan t and herbivores feeding on it (Yang and Rudolf 2010) , and potentially change what crop or pest management practices should be followed in the wake of a heat wave. For example, soybeans and corn responded differently to heat stress in certain life stages than others (Siebers et al. 2015, 2017) . Heat stress also had varying effects on different life stages of aphids, with younger nymphs more susceptible than older in stars and adults ( Chiu et al. 2015 ; Zhang et al. 2015 ) . However, previous studies often did not incorporate the daily variation in temperature between night and day, which is key to linking research results to practical predictions (Dowd et al. 2015). I examined the impacts of heat waves and the ir timing on a crop insect pest interaction using both laboratory and field experiments that exposed multiple crop life stages to heat wave conditions in the presence or absence of herbivores. I did this using t he specialist insect herbivore CPB and its ho st plant, the potato . I used growth chambers to create heat waves in the laboratory and open top chambers fitted with ceramic heaters to simulate heat waves over field - grown potatoes. I aimed to identify the potato life stages that are most susceptible to heat stress and how beetle survival and weight are impacted by feeding on heat - stressed plants . 7 METHODS Experimental o rganisms For both lab and field experiments, we grew potato plants from certified seed potatoes ( Solanum tuberosum r to planting, seed potatoes were cut and allowed to suberize overnight for the growth chamber experiment and for 7 d for the field experiment (~40 g per seed piece for lab experiments and ~64 g for field experiment). For all experiments, we obtained CPB f individuals collected from a potato field in Stanton, Michigan in 2017 and kept in continuous culture in the l aboratory (16L:8D, 23ºC). Growth chamber experiment In a growth chamber experiment ( 0 1 /201 8 0 6 /2018 ) I investigate d the effects of heat stress and its timing on potatoes, larvae, and their interaction. This experiment exposed 156 potato plants to either constant control conditions, an early heat wave, or a late heat wave (52 per heat wave treatment), allowing me to exa mine the effects of heat wave timing. I added three 1st instar beetle larvae to half of the plants in each treatment, making six total treatments with 26 plants per treatment. The larvae were added all at the same time, after the early heat wave but before the late heat wave, so that some larvae fed on plants that already experienced a heat wave, while another group of larvae fed on plants during a heat wave. This allowed the determination of direct and indirect effects of heat stress on the larvae and thei r interaction with the plants. I started the experiment by planting 200 potato seed pieces into 450 ml round plastic pots filled with soil (Suremix perlite, Galesburg, MI) and placing them into a growth chamber (BioChambers Inc., Model FxC19; Winnipeg, Can ada). This chamber the constant conditions 8 control chamber was set to 16L: 8D photocycle, 26ºC during the day, and 14ºC at night, which are the average June temperatures for Southern Michigan (Coldwater , MI weather station; MSU Enviroweather). Light levels were measured at bench level and set to 170 µmol m - 2 s - 1 using a light meter (Apogee Instruments, Inc., Logan, UT). I used a 2 h linear ramp between day and night temperatures to prevent sudden heat sho cks, resulting in 12 h of day and 8 h of night tempera tures. Humidity was set at 40% and fan speed was at 75%. When plants were 2 weeks old, we selected 156 fully emerged plants of similar size to be used in the growth chamber experiment. Trays of plants w ere rotated 180 degrees and to different benches withi n the chamber on watering days to avoid uneven conditions. All plants were watered three times per week (25 ml of water/plant). I fertilized once per week with 25 ml of 750 ppm 20 - 20 - 20 N - P - K solution ( . When plants were 45 d old in the vegetative growth phase, I applied the early heat wave treatment on 51 plants by moving them from the control chamber to a second chamber set to heat wave conditions. This chamber was set to simulate heat waves with a 40 ºC day maximum and a 26ºC night minimum. I chose these conditions because an analysis of 1897 2016 temperature data from Michigan (Coldwater, MI weather station) indicated such an event was in the 99th percentile of past events, and these rare events are predicted to become 8 20 times more likely after 3°C of mean warming (Fischer and Knutti 2015). The heat wave lasted for 4 d, after which plants were moved back to the control chamber. Neither control no r treatment plants receive d fertilizer during heat waves to minimize negative effects of the fertilizer during a heat stress event (Slafer & Savin 2018) . When plants were 56 d old, 7 d after the end of the early heat wave treatment, I placed three 1st inst ars on half of all the plants in control, early heat wave, and late heat wave 9 treatments. This delayed placement allowed us to detect the lasting impact of the early heat wave on plant nutritional quality while not exposing those larvae to the direct heat stress. Conversely, the larvae in the late heat wave treatment did experience the extreme heat conditions. Larvae were c ontained within bags ( California Containers, 19 L, 200 - micron white Nylon mesh material) placed over each plant . A 38 - cm bamboo stake wa s placed in the soil to support the bag and the plant ; the bag was twist - tied closed around the base of the plant and bo ttom of the stake. Bags and stakes were also placed over all control plants which did not receive any larvae. Only egg clutches large en ough to contribute 3 larvae to each treatment were used. The clutch number that the larvae came from was randomly assign ed to the plants. I applied the late heat wave treatment when plants were 59 d old and in the tuber initiation stage by moving another 51 plants from the control chamber to the heat wave chamber for 4 d. This was 3 d after beetle larvae were added to plan ts, which means that larvae and plants experienced heat wave conditions together. After the late heat wave, plants and their larvae were moved back to the control chamber. Larvae were left on plants for another 3 d before being removed. All larvae were on plants for a total of 10 d which is the minimum time from egg to pupation at (Ferro et al., 1985) . On d 66, larvae were removed and larval survival, mass, and instar were recorded. Plants were destructively sampled to obtain tuber mass 70 d after planting. Field experiment I examined the effects of heat wave timing on potatoes, CPB larvae, and their i nteraction using a field experiment at the Michigan State University Kellogg Biological Station (Hickory Corners, MI). I generated heat wave conditions in field plots using ceramic heaters and open - top chambers at two di fferent times in the growing season for two different age classes of plants. 10 In the field, I set up 96 1.5 x 1.5 m plots. Each plot had 10 potato plants total, 5 in each of two rows, which were planted 14 d apart ( 05/25/2018 and 06/08/2018 ) to obtain two age classes of plants. The field had 0.76 m spacing between rows and 0.25 m between seed pieces within a row. I randomly assigned two plants from each row , and or age class, within each plot to a beetle treatment or a no - beetle control. Fin ally, I randomly assigned each plot to a July heat wave (n= 24), or an August heat wave (n= 24). This design resulted in 12 treatment combinations of heat wave timing, beetle presence, and plant age (Table 2. 1). Three d before the start of each heat wave, plants assigned a beetle treatment were covered with a 19 L mesh bag with five 1st instars per bag. Plants in each paired no - beetle control were simultaneously bagged without larvae. We began simulating heat wave conditions on plots assigned to heat wave t reatments 3 d af ter larvae were placed on plants, ( 07/17/2018 , 08/13/2 018 ) for each timing treatment and continued treatment for 96 h. At the conclusion of each heat wave, we removed the chambers and left the larvae on the plants for another 3 d. After a total of 10 d on the plants the larvae were removed and counted. The survival, mass, and instar of each larva were recorded and the amount of insect defoliation was visually estimated. On ( 09/22/2018 ) , the four experimental plants in every plot we re dug up by hand and pl aced in paper bags. These potatoes were transported to the laboratory and stored at 2.8°C. They were assigned a scab rating from 0 - 5 (0 = no scab and 5 = > 50% of tuber covered with pitted lesions) and sorted into standard potato si ze classes (USDA, 1997). Total tuber mass was recorded for each plant, then potatoes with deformities were removed and healthy potatoes were counted and weighed to assess the marketable yield of each size class. 11 Heat wave methods The heat waves were creat ed by using open - top cha mbers and ceramic heaters (300 - watt, Tempco, Inc., IL). The chambers were constructed of wood and anti - condensate greenhouse plastic (6 mil, 91% light transmittance, Poly - Ag Corp., CA). The chambers were 1.5 by 1.5 m, 1.2 m tall, wi th a 0.6 m 2 square openi ng at the top. These chambers were fitted with ceramic heaters that were turned on at 2100h and off at 0900h to maintain elevated plant canopy temperatures at night. Leaf temperatures were recorded for a subset of plants (digital in frared thermometer, Fluk e Corporation, Everett, WA) on the 3rd d of the experiment between 1200h and 1300h and the 4th d from 0400h to 0600h. The soil surface temperature and soil temperature 10 cm below the surface were recorded at the same time. Ambient air temperatures were re corded every 15 min (HOBO Pendant data loggers in radiation shields, Onset Computer Corporation, Bourne, MA) in four heat - treated plots and in four control plots for the duration of each heatwave. Two of the data loggers recorded humidity (HOBO Pro v2 dat a loggers in radiation shields, Onset Computer Corporation , Bourne , MA ). In order to calculate the average temperatures for the two treatments, we selected two 3 - hr time periods (0300 - 0600h, 1300 - 1600h) to represent night and day temperatures, respectively . The heat wave treatment elevated air temperatures in the July experiment by 8.7°C to 43.0°C (+/ - 0.1°C) during the day and by 3.5°C to 21.4°C (+/ - 0.4°C) at night, relative to the control (day and night temperatures av eraged over 3 - h block in middle of day and night). Potato leaf temperatures increased by 3.6°C during the day and 8.4°C at night, relative to the control (Table 2. 2). In the August experiment, air temperatures were elevated by 4.0°C to 38.3°C (+/ - 0.5 °C) during the day and by 4.7°C to 24.7 °C (+/ - 0.3°C) at night (see Table 2. 2 & 2. 3 for details). Leaf temperatures were elevated by 1.4°C during the day and 6.8°C at nigh t, relative to 12 the control (Table 2. 3 ). The lower increase in temperature during the August experiment was caused by slightl y different weather conditions during the August experiment . Statistical analyses I analyzed the effects of heat waves on beetle survival and rate of development to 4th instar using generalized linear mixed - effects models with a binomial distribution. I a nalyzed tuber mass and beetle mass using a linear mixed effects model. In the growth chamber experiment, I included larval clutch as a random effect to account for potential correlations among larvae from the same clutch. In the field experiment, I include d plant age during the heat wave treatment (the ontogenetic heat wave timing) as a pr edictor in all models. For potato yield, I also included the beetle treatment and its interaction with ontogenetic heat wave timing as predictors. Our models accounted for correlations among responses from plants within the same plot using a plot - level ran dom effect. I repeated analyses for both the July and August field experiments. I conducted all analyses in the lme4 and glmmTMB packages in R (Bates et al., 2015, R Core Team 2019). I used likelihood ratio tests to compare models with and without each of our fixed - effect predictors (Bolker , 2009). When treatment effects were significant ( = 0.05), em package. 13 RESULTS Growth chamber experiment B eetle performance The environmental chamber heat wave treatments significantly reduced CPB larval survival ( 2 =23.1, df=2, P<0.01; Fig. 2. 1 . A). Larvae that experienced the late heat wave had 48 % lower survival than larvae on control plants (z =4.32, P<0.01). Larvae in the early heat wave treatment, in which larvae were placed on plants 7 d following the conclusion of heat wave temperatures, had 20% lower survival relative to controls (z =2.13, P =0.03), indicating that the heat wave treatment indirectl y reduced beetle survival by altering plant quality. In contrast, neither heat wave timing treatment significantly influenced beetle larval mass ( 2 =0.67, df=2, P=0.71) or larval instar development rate ( 2 =0.32, df=2, P=0.85), suggesting that larvae tha t survived a heat wave grew normally (Fig. 2. 1 . B ). 14 Figure 2.1. Survival and mass of Colorado potato beetle larvae feeding on plants exposed to a n early heat wave and late heat wave treatments. Blue circles denote larvae feeding on control plants and red triangles represent larvae feeding on plants exposed to a heat w ave treatment. ( A ) Larval survival was lowest in the late heat wave treatment, in which larvae were present on plants during the heat wave. Larval survival was slightly reduced by the early heat wave treatment in which the heat wave occurred on plants one week prior to larval addition. ( B ) Larval mass did not differ between control and heat wave treatments. Points and error bars show means +/ - one standard error. Means with different letters are significantly different from each other ( Statistical test p<0.05) Potato yield The environmental chamber heat waves signi ficantly reduced potato tuber mass (F 2, 154 =17.94, P<0.01 ; Fig. 2 ), and these effects depended on heat wave timing. The early heatwave (45 d after planting) reduced tuber mass by 87% relative to the controls (t 154 = 5.95, P < 0.01). The late heat wave trea tment (59 d after planting) led to a 35% reduction in tuber mass (t 154 = 5.95, P = 0.02). There was no overall effect of larva l presence on tuber mass in this experi ment (F 1, 155 = 0.23, P = 0.63), nor was there a significant interaction between the heat w ave treatments and 15 larval presence (F 1, 151 = 0.22, P = 0.8). However, there was a trend for CPB presence to have more negative effect on yield when combined with he at waves, especially the late heat wave (Fig. 2 .2. ). In the late heat wave, plants with CPB produced 43% lower yield than control plants without larvae (t 151 =2.02, P=0.075), a substantially larger reduction than the only 26% decline in yield as result of the heat wave alone compared to control plants (t 151 =1.20, P = 0.26). Figure 2. 2. Averag e tuber mass produced per plant for all heat wave treatment combinations after 70 d. Blue circles denote control plants and red triangles represent plants exposed to a heat wave treatment. The early heat wave had the greatest negative effect on tuber mass. The addition of 3 beetle larvae had no effect on yield compared to plants without larvae and of the same heat wave treatment. The addition of 3 beetle lar vae to the late heat wave treatment did decrease tuber production compared to control groups. Points and error bars show means + / - one standard error. 16 Field experiment B eetle performance In the July block, the heat wave treatment decreased larval survival by 25% regardless of plant age ( 2 =8.94, Df=1, P=0.004). There was a 19% reduction in larval survival for those experiencing a heat wave and feeding on 6 - week - old plants (Z=2.8, P=0.02) and a 31% reduction for larva experiencing a heat wave and feeding on 8 - week - old plants (Z=2.8, P=0.0 2; Fig. 2. 3 . A) compared to survival of larva feeding on non - heat wave 6 and 8 - week - old control plants respectively. Age of plants a lone only accounted for a 6% decrease of larval survival for those feeding on 8 - week - old plants compared to 6 - week - old plants ( 2 =0.60, Df=1, P=0.44). In contrast, in the August block, heat wave treatment was not a significant predictor, only causing a 5 % reduction in larval survival ( 2 = 0.27, Df= 1, P= 0.61). However, plant age at time of larval placement was an important fa ctor for larval survival with 12 - week - old plants having a 27% lower survival compared to larvae on 10 - week - old plants ( 2 = 9.01 , Df = 1, P = 0.003; Fig. 2. 3 . B). Larvae feeding on 12 - week - old control plants had a 17% decrease in larval survival compared t o those feeding on 10 - week - old control plants (Z = 3.04, P = 0.007). The impact of the heat wave in August on larval survival var ied drastically depending on the age of plants they were feeding on. Larvae feeding on 10 - week - old plants and experiencing a he at wave saw a 6% increase of larval survival (Z = - 0.57, P = 0.57). In contrast, larval feeding on 12 - week - old plants and experien cing a heat wave had a 19% reduction in larval survival (Z = 1.25, P = 0.26). Larvae feeding on 12 - week - old plants during the A ugust heat wave treatment had 37% lower survival than larvae feeding on 10 - week - old heat wave plants (Z = 3.04, P=0.01). This sug gests that older potatoes interact with the heatwave in a way that decreases palatability or nutrition to cause lower larval su rvival rates. 17 Figure 2 . 3. Average survival of Colorado potato beetle larvae feeding on different age plants for (A) July and (B) August heat wave treatments in a field at the Kellogg Biological Station. Blue circles denote control plants and red triangl es represent plants exposed to a heat wave treatment. Larval survival was lower for those that experienced a July heatwave regardless of plant age. There was no difference in survival of larvae between control and heatwave treatments for the August heat wa ve, but the larvae that were feeding on younger 10 - week - old plants and experienced a heat wave had a much higher survival rate than those that were feeding on the 12 - week - old plants and experienced a heat wave. In the July block, regardless of plant age, larval mass increased by 13 % in heat wave treatments compared to larvae feeding on no - heat wave control plants ( 2 = 3.89, df = 1, P = 0.049; Fig. 2. 4 . A). Average larva l mass increased by 5% for larvae feeding on 6 - week - old plants exposed to heat wave tr eatments (t= - 0.509, df=80, P =0.61) and by 21% for larvae feeding on 8 - week - old plants in the heat wave treatment (t= - 2.304, df=80, P =0.14) compared to larvae feeding on control plants of the same age. Plant age alone did not have an impact on larval mass i n the July block ( 2 = 0.07, df = 1, P = 0.79). The greatest impact on larval mass occurred in larvae that 18 fed on 8 - week - old potato plants and experienced a July heat wave, with 21% higher mass than larvae that fed on 8 - week - old non - heat wave control plants. In the August block, larval mass was not impacted by the heat wave treatment ( 2 = 1.87, df = 1, P = 0.17; Fig. 2. 4 . B), or plant age ( 2 = 2.68, df = 2, P = 0.26). Figure 2. 4. Mean mass of Colorado potato beetle larvae feeding on different age plants exposed to (A) July and (B) August heat wave in a field at the Kellogg Biological Station. Blue circles denote control plants and red triangles represent plants exposed to a heat wave treatment. Larval mass was 30% hi gher for larvae that fed on 8 - week - old potatoes and we re exposed to a July heat wave (P<0.05) compared to non - heat wave controls. Heat waves significantly influenced larval mass when modeled collectively ( 2 =6.0094, df=6, P= 0.01423). 19 Potato yield O lder plants (those planted in May vs June) had 27% higher total yield ( 2 = 50.038, df = 1, P< 0.01) and 46% higher marketable yield then younger plants regardless of heatwave or beetle treatments ( 2 = 53 .34, df = 1, P< 0.01). Because marketable tuber mass is most important to production that i s what we focus on here . In the July experiment, plant age was the most important predictor of marketable tuber mass with older 8 - week - old plants having a 47% higher marketable yield than 6 - week - o ld plants ( 2 =23.56, df=1, P<0.001). The beetle treatment and heatwave treatment had no effect on marketable tuber yield as single factors ( 2 =0.25, df=1, P=0.62; 2 =2.57, df=1, P=0.11 respectively). However, a model with an interaction between plant ag e and heat wave was found to be significantly better than the model with plant age alone ( 2 =6.7, df=1, P=0.035). We also found 8 - week - old plants that experience a heat wave had a 28% decrease in marketable yield (t=1.97, df=182, P=0.12), and plants that experienced both heat wave and beetle treatments have 41% lower marketable yield than plants in the control treatment (t= - 2.8, df=182, P=0.02 ; Fig. 2. 5 . A). In the August experiment, plant age was the only significant predictor of marketable tuber mass with older 1 2 - week - old plants having a 45% higher marketable yield than 10 - week - old plants ( 2 =27.1, df=1, P<0.001; Fig. 2. 5 . B). The beetle treatment and heatw ave treatment had no significant effect on marketable tuber yield as single factors ( 2 =0.4, df=1, P=0.52; 2 =0.2, df=1, P=0.65 respectively). 20 Figure 2. 5. Blue circles denote control plants and red triangles represent plants exposed to a heat wave treatment. (A) Heat wave treatment significantly reduces marketable tuber mass in 8 - week - old plants that experienced a July heat wave. (B) Marketable tuber mass in the August heat wave was si gnificantly influenced by plant age but heat wave and beetle trea tments had no effect. Blue circles denote non heat wave controls and red triangles denote heatwave treatments. DISCUSSION I examined the impact of heat waves and heat wave timing on insect - plant interactions on potato using an environmental chamber experiment and a field experiment . The chamber experiment revealed that the timing of both heat wave and herbivory is important: heat waves significantly reduced herbivore survival regardless of plant age and whether the larvae experienced the heat waves directly, although the greatest reduction in survival occurred when herbivores were directly exposed to the heat wave. This indicates that the direct effect of heat waves is not necessary for herb ivores to experience negative effects and that these events can 21 have both direct and indirect effects, possibly leading to an emergent effect on herbivore survival. It is interesting to note that larval mass and instar was similar across control and heat w ave treatments; this could be due to behavioral chang es whereby surviving larvae were able to locate microclimates on the plant that allowed them to temper the effects of the extreme heat as observed in spiders and other beetles (Barton & Schmitz, 2009;War d & Seely, 1996) . This similarity in mass and development is perhaps best explained by overlaying average temperatures experienced by larvae in control ( ) and heat wave ( ) treatments . These two groups of larva e both experienc ed similar average temp eratures at the op timal CPB developmental (Ferro et al., 1985) . Due to the experimental set up, we cannot conclusively distinguish between the effect of plant age and direct/indirect larval heat wave exposure on larval survival, but our results indicate that the effects of heat stress on potato persist in the leaf tissue for at least a week after a heat wave. Direct heat wave effects on undamaged plants were more pronounced when the plant s were younger, indicating that younger potatoes a re more sensitive to stress than older plants. In the field, heat wave effect s on larval survival depended on heat wave timing, with more consistent negative impacts earlier in the season on younger plants and variable effects on older plants later in the season. It is likely that heat waves still reduced palatability and nutrient v alue in older plants but effects on survival are not as drastic. L arval mass however, was consistently higher on heat waved plants regardless of their age. This over compensatory response is potentially caused by larvae having higher metabolic rates and co nsuming more plant tissue during heat waves, although we did not measure leaf area consumed (Neven, 2000) . The difference in patterns of larval weight between the laboratory and field experiments could be due to the fact that larvae experiencing 40 ° C for 14 h in the environmental chamber was near their 22 thermal maximum in but in the field experiment heat waves had similar high temperatures with a shorter duration of high heat makin g average temperatures closer to developmental optimum. Heat waves also had negative impacts on potato yield but only for plants in the early reproductive life stages of tuber initiation and e arly tuber bulking (8 - wk - old). Potatoes are most sensitive to te mperature extremes during the tuber initiation phase, a nd are less responsive once the tubers are formed, during tuber bulking (Kim and Lee 2019) , therefore it is possible that the lack of consistent negative yield response was because some plants were sti ll in the vegetative stage (6 - wk - old) or had already started bulking (10 & 12 - wk - old) . With heat waves in the Midwest occurring most commonly from early June to mid - August the risk of these sensitive life stages of potato being exposed to stressful tempera tures is high, especially for later crop plan tings (Anderson et al., 2011) . Extreme heat affects numerous aspects of plant - herbivore relationships, but a decrease in herbivore survival is a frequent outc ome (Sentis et al., 2003, Ferro et al., 1985, Chen et al., 201, Logan et al., 1985). W hile heat waves are relatively brief, the stress or heat - shock they cause can persist in plants and affect herbivores indirectly after the event as demonstrated with the lab experiment (Al - Whaibi, 2011) . Our study highlighted that heat waves on potatoes at different developmental stages could differ in how the heat w ave impacts them and require different management responses. Studying extreme events therefore requires the adoption of a temporally - explicit framework that links seasonal variation, organismal ontogeny, and event timing to ecological outcomes. In addition , heat waves may speed up herbivore development with significant negative impacts on crop production such as an extra herbivore generation is possible earlier than predicted by the gradual warming models. I n this system , similar to other crops , the most sensitive life stage of potato to heat waves and herbivore stress are the early reproductive 23 stages suggesting management practices such as irrigation or pesticide applications should be diligently used to mitigat e stressful conditions during this sensitive growth stage (Kim and Lee 2019; Siebers et al. 2015). Research that manipulates these various stressors and the timing at which they occur is key to informing an adaptive management strategy and should continue to be investigated in the future. 24 APPENDIX 25 Table 2. 1 The number of replicates by month of planting, heat wave timing and presence or absence of Colorado potato beetle larvae in a field experiment at the Michigan State University Kellogg Biological Station in 2018. We generated heat wave conditions in field plots using ceramic heaters and open - top chambers each with 10 po tato plants (5/month). May June 0 larvae 5 larvae 0 larvae 5 larvae Control 44* 44 44 44 July Heat Wave 22 22 22 22 August Heat Wave 22 22 22 22 *Controls were paired within blocks with heat treatments, resulting in twice as many control plants as heat treatment plants. Table 2. 2 Mean ± SEM temperatures (°C) for 3 - hour increments during heat of the day and cool of night during field heat wave experiments in Michigan in an experimental potato field in 2018. Heat waves were created using open top chambers and ceramic heaters. Jul y August Control Heat wave Control Heat wave Day Night Day Night Day Night Day Night Air temperature 34.3±0.8 17.9±0.1 43.0±0.1 21.4±0.4 34.3±0.5 20.0±0.03 38.3±0.5 24.7±0.3 Leaf surface 28.8±0.3 17.3±0.1 32.4±0.3 25.7±0.5 24.9±0.1 18.9±0.1 26.3±0.2 25.7±0.2 Soil surface 33.3±0.3 19.3±0.2 35.8±0.4 25.7±0.3 27.0±0.2 20.8±0.3 28.7±0.3 25.8±0.3 Soil sub - surface 27.3±0.2 24.0±0.1 29.4±0.2 26.1±0.2 25.9±0.2 23.4±0.1 27.1±0.2 25.4±0.2 26 Table 2.3 Temperature (°C) differences between heat wave and control treatments in open top chambers for heat waves created in an experimental potato field in Michigan in 2018. July August Day Night Day Night Air temperature 8.7 3.5 4 4.7 Leaf temp 3.6 8.4 1.4 6.8 Soil surface 2.5 6.4 1.7 5 Soil sub - surface 2.1 2.1 1.2 2 27 CHAPTER 3 : Consequences of heat waves on plant, herbivore, disease interactions ABSTRACT Heat waves, b rief periods of intense heat and decreased pre cipitation , are becoming more common globally. The immediate negative effects of heat stress on organisms are well understood, but we have a poor understanding of the potentially lasting ecological effects of heat waves on biotic interactions among plants, insect her bivores, and pathogens . An understanding of how heat waves alter the dynamics of ecological interactions in agricultural systems would improve our ability to manage crop pests in an increasingly variable climate. I examined how heat waves influe nce the growth and development of potatoes and two potato pests: CPB and potato early blight ( A. solani ) . I inoculated field grown potato plants with early blight and CPB 3 d prior to simulating heat wave conditions in the field using open - top chambers and ceramic heating elements . Heat waves and early blight treatments had no effect on CPB survival, but larval mass was 16% and 44% higher for larvae feeding on blighted plants and plants exposed to heat waves respectively . Larvae developed faster in heat wave treatments suggesting a n increased likelihood of an additional generation of the pest. Potato yield was not affected by any single stressor, b ut when heat waves were added to plants with early blight and CPB total yield was reduced by 1 4 % and 17% respectively. These results sugg est that multi - species interactions are heavily influenced by short duration, high intensity heat waves and the projected increase in frequency of these extreme events justifies future research on their ecological effect to improve agricultural sustainabil ity and food security 28 INTRODUCTION Most climate change research has focused on the gradual increase in mean temperature predicted to occur in the next 30 - 100 years, neglecting the inherent variability of extreme weather such as heat waves (Lloret et al. 2012; Parmesan and Yohe 2003 ) . The current and predicted increase in the frequency and intensity of extreme weather events has become a major focus of climate change research (Meehl and Tebaldi 2004 ; Rahm storf and Coumou 2011 ; Fischer and Knutti 2015 ; Perkins - Kirkpatrick and Gibson 2017 ). Extreme events, such as heat waves, cold snaps, drought s , and floods, are localized abiotic disturbances that have direct negative effects on organisms but also disrupt t rophic interactions (Barlow et al. 2015) . These stressors have directly reduced crop yields and caused phenological mismatches of plants and insects altering community composition for years afterwards which is of ecological significance (Bisbis et al. 2018 ; Lesk et al. 2016; Smith 2011; Tho mpson et al. 2013) . T emperature influence s the development and life cycle of all organisms differently as each ectothermic organism has its own thermal optimum, minimum and maximum for growth and reproduction. While heat waves have direct effects on insec ts, disease s , and their host p l ants, the impact on the interaction among them is much less studied ( Szczepaniec and Finke 2019 ). Increased temperatures can increase development rates for insects, diseases, and plants whic h can disrupt the synchronicity of tritrophic interactions ( Szczepaniec and Finke 2019 ; Dyer et al. 2013) . Due to the difficulty of replicating heat waves under field conditions, most research on heat wav es to date has been done in growth chambers or greenhouses limiting their ability to let ecological interactions play out (Brown et al. 2004; Harrington et al. 1999 ; Parmesan and Yohe 29 2003) . Moreover, m any of these studies are conducted at constant tempera tures that do not capture key daily variation that occurs during a heat wave (Lloret et al. 2012) . The goal of this research was to improve our understanding of how heat waves alter multi - species interactions in a crop system , thereby enhancing the sustai nability of agric ulture in the face of an increasingly variable climate . I addressed this by using open - top chambers and ceramic heaters to generate heat wave conditions on crop plants , an insect herbivore, and a common fungal disease in the field . I used a fully - crossed experimental design to address how extreme heat affects the fitness of potato, the survival and growth of CPB , and the severity and incidence of a common fungal disease, potato early blight ( A. solani ). I addressed the following questions: (1) How do heat waves and early blight influence CPB growth and survival? (2) How do heat waves and CPB impact early blight infection rate and severity? (3) How do heat wave s , CPB, and early blight interactions influence potato yield? METHODS Experimenta l o rganisms Insect I use d CPB and early blight because they are both common pests of potato plants, occur at the same time during the season in potato fields, and are hypothesized to have synergistic effects on one another (Tsedaley, 20 14 ). We obtained CP B from a lab colony reared on potatoes Michigan in July 2019 and kept in the laboratory at a light period of 16 h light :8 h dark and constant temperature at 23 º C. Clutches of eggs were collected over 2 d to obtain enough insects 30 to place 15 larvae per plot, for a total of 1,200 larvae for the experiment. All larvae were randomly mixed together in a container before selecting 15 individuals to be added to each pl o t. Disease An isolate of Alternaria solani was obtained from a Michigan State University collection (isolate A00219; collected Sanilac Co., MI in March 2019). The isolate was originally collected from a symptomatic potato tuber, variety Onaway. An approxi mately 0.5 cm3 piece of tissue was excised and surface disinfested by submerging for 1 min in 0.6% sodium hypochlorite followed by rinsing in sterile deionized water and blotting on sterile filter paper. The disinfested tissue was plated on 1.5% wat er agar and incubated at 21 - 24 °C for 3 - 5 days. Resulting mycelial growth and conidial sporulation was morphologically characterized as A. solani. For molecular identification, fragments of the internal transcribed spacer (ITS) were amplified using the ITS 1 and I TS4 primers (White et al. 1990) and sequenced; resulting sequences were matched with known sequences in GenBank using the BLAST algorithm. Characterized growth was hyphal tipped onto clarified V8 agar (Rosenzweig et al. 2008) 16 g L - 1 agar, 3 g L - 1 calci um carbonate, and 0.05 g L - 1 benomyl amended with 30 µg ml - 1 streptomycin. The isolate was stored on sterile filter paper at - 20 °C to preserve viability. Active cultures of the isolate were maintained on clarified V8 agar amended with 30 µg ml - 1 strepto mycin, as described previously. After 10 - 14 days of colony growth at 21 - 24 °C under a 17 h light: 7 h dark regime, conidia were harvested by gently scraping the surface of the sporulating colony with a microspatula. Collected spores were diluted wit h deion ized water to achieve a conidial spore suspension at a concentration of ~1x10 4 spores/ml with 10% Tween 20 (v/v) added to act as a surfactant. Plants were infected with this solution using a hand sprayer by 31 spraying 25 ml of inoculum, which was unti l it dr ipped off the leaves. Non - blight treatments received the same treatment with distilled water and the surfactant but without the spores. Plants were sprayed in the last two hours of daylight on (08/01/2019) . Plant For this field experiment, I grew potato pl ants from certified seed potatoes ( Solanum tuberosum - to the Kellogg Biological Sta tion organic fields and planted on (06/12/2019) . Experiment al desi gn I investigated the effect of heat stress in the form of a 5 - d heat wave on the interaction among A. solani , CPB larvae and potato using a full - factorial field experiment at the Michigan State University Kellogg Biological Station (Hickory Corners, MI) . I generated heat wave conditions in field plots using ceramic heaters and open - top chambers. In the field, I set up 130 1.5 x 1.5 m plots on June 12 2019. Each plot had 4 potato plants with 2m spacing between rows and 0.25 m between seed pieces within a row. I randomly assigned the two center plants in each plot to a beetle treatment or a no - beetle control. Final ly, I randomly assigned each plot to a control treatment (n= 20), early blight treatment (n= 20), a heat wave treatment (n= 20), or a n early bli ght + heat wave treatment (n= 20). This design resulted in 8 treatment combinations of early blight , heat wave, and beetle. The morning after plants were inoculated with early blight, 3 d prior to start of the heat wave, plants assigned to a beetle treatme nt received 15 1st instar larvae and were covered with a 19 L mesh bag and secured with a twist tie around the b ase of the stem. The no - beetle control 32 plant in each plot was bagged in the same manner at the same time without larvae. I began simulating heat wave conditions on plots assigned to heat wave treatments 3 d after larvae were placed on plants (08/05/2019) a nd continued for 120 h. At the conclusion of the heat wave , open - top chambers were removed on (08/10/2019) and the larvae were left bagged on t he plants for another 2 d. After a total of 10 d on the plants the larvae were removed and counted on (08/12/201 9). The survival, mass, and instar of each larva were recorded and the percent of insect defoliation was visually estimated. Plant height, % dea d tissue, and % defoliation was recorded weekly up until harvest (omitting the first week of September). On (09/14/2019) , the potatoes produced by the two experimental plants in every plot were harvested by d igging them up by hand and plac ing them in pape r bags. These potatoes were transported to the laboratory and stored at 2.8°C for approximately three months. They were then removed and assigned a scab rating from 0 - 5 (0 = no scab and 5 = > 50% of tuber covered with pitted lesions) and sorted into standa rd potato size classes (USDA, 1997). Total tuber mass was recorded for each plant, then potatoes with deformi ties and rot were removed and healthy potatoes were counted and weighed to assess the marketable yield of each size class. Results were similar for total tuber yield and marketable yield, so I focus on total yield below. Heat wave methods The heat waves were created by using open top chambers and ceramic heaters (300 - watt, Tempco, Inc., IL). The chambers were constructed of wood and anti - condensate g reenhouse plastic (6 mil, 91% light transmittance, Poly - Ag Corp., CA). The chambers were 1.5 by 1.5 m at the base, 1.2 m tall, with a 0.6 m 2 square opening at the top of the plastic sides. These chambers 33 were fitted with ceramic heaters 90 cm above the soi l surface that were turned on at 2100h and off at 0900h to maintain elevated plant canopy temperatures at night. Heaters were turned on outside of those hours when there was greater than 50% cloud cover f or a minimum of 30 minutes, as the chambers increase d solar radiation during the day to elevate the ambient air temperature. Leaf temperatures were recorded for a subset of plants (digital infrared thermometer, Fluke Corporation, Everett, WA) on the 4 th d of the heat wave between 0400h and 0600h and the 4 th d from 1200h to 1400h. The soil surface temperature and soil temperature 10 cm below the surface were recorded at the same time. Ambient air temperatures were recorded every 15 min (HOBO Pendant data lo ggers in radiation shields, Onset Computer Corporati on, Bourne, MA) in four heat treated plots and in four control plots for the duration of each heatwave. Two of the data loggers recorded humidity (HOBO Pro v2 data loggers in radiation shields, Onset Com puter Corporation). In order to calculate the averag e temperatures for the two treatments, I selected two 3 - hr time periods (0300 - 0600h, 1300 - 1600h) to represent night and day temperatures, respectively. The heat wave treatment elevated ambient air tempera tures by 6°C to 38.1°C (+/ - 0.6 SE) during the day a nd by 3.9°C to 21.1°C (+/ - 0.2 SE) at night, relative to the control. Potato leaf temperatures increased by 5.1°C during the day and 9.6°C at night, relative to the control (Table 3.1 and 3.2 ). Statistical analyses I conducted all analyses in the lme4 pac kage in R (Bates et al., 2015, R Core Team 2019). I used likelihood ratio tests to compare models with and without each of our fixed - effect predictors (Bolker 2009). When treatment effects were s ignificant ( performed to determine I analyzed tuber mass and beetle mass using a linear mixed - effect models with a normal distribution , and 34 survival and development to 4 th instar using generalized linear mixed - effects models with a binomial distribution . The models accounted for correlations among responses from plants within the same plot using a plot - level random effect , and correlations among CPB on the same plant with plant and plot level random effects. For potato yield, we incl uded the beetle treatment and its interaction with early blight as predictors. RESULTS Beetle performance Larval survival Early blight did not affect larval survival ( 2 =1.13, df=1, P =0.29), nor did the heat wave treatment ( 2 =0.2, df=1, P =0.66) . The com bined effect of early blight and heat wave on larval survival was also not significant ( 2 =1.34, df=1, P =0.51; Fig. 3.1. ). 35 Figure 3.1 . Mean (+/ - SEM) survival of Colorado potato beetle larvae feeding on control plants (blue dot), early blight inocula ted plants (brown dot), heat wave (HW) plants (red dot), and plants exposed to both early blight and heat wave treatments (yellow dot). Larval mass Larvae exposed to heat wave treatments had a 44% higher larval mass than those feeding on control plants (t = - 4.5, df= 80.3, P <0.01; Fig. 3.2. ). Larvae feeding on blighted plants and exposed to heat wave treatments also had a 43% increase in mass compared to control (t= - 4.67, df= 77.6, P <0.01) and 25% increase compared to those feeding on blighted plants (t= - 2.9 5, df= 74.4, P =0.02). Larvae feedi ng on early blight inoculated plants had a 15% increase in mean mass compared to those feeding on control plants (t= - 1.6, df=74.5, P =0.37). Any positive effect of early blight on larval mass appears to be canceled out by e ffects of the heat wave, demonstra ted by a comparison between larvae in the heat wave treatment compared to those in the early blight and heat wave treatment (t= - 0.12, df=80.3, P =0.99). 36 Figure 3.2 . Mean mass of Colorado potato beetle larvae feeding on co ntrol plants (blue dot), early blight inoculated plants (brown dot), heat wave (HW) treated plants (red dot), and plants exposed to both early blight and heat wave treatments (yellow dot) in a field at the Kellogg Biological Station. Larval development T he average larval instar reached was most influenced by whether the larvae experienced a heat wave or not, with 5% more larvae achieving the final 4th instar in the heat wave treatment (z= - 3.1, df=inf, P =0.01) compared to control. The larvae experiencing e arly blight and the heat wave treatment had a similar proportion of 4 th instars as control treatments (z= - 2.1, df=inf, P =0.13), but 4.4% more than those feeding on just blighted plants (z= - 3.1, df=inf, P =0.01 ; Fig . 3.3. ). 37 Figure 3.3 . Proportion of surviv ing larvae that achieved 4th instar feeding on control plants (blue dot), early blight inoculated plants (brown dot), heat wave (HW) treated plants (red dot), and plants exposed to both early blight and heat wave treatments (yellow do t). Larvae were placed on plants as first instars and removed after 10 d , with the heat wave treatment occurring from day 3 to day 8. Disease development Di sease prevalence , calculated as the cumulative sum of percent leaf with disease, was significantly higher on plants inoc ulat ed with the A. solani solution ( 2 =79.6, df=1, P <0.01) , indicating that our early blight treatment was successful (Fig. 3.4. ) . The best model of disease prevalence including an additive effect of the early blight treatment and the heat wave treatment ( 2 =8.9, df=1, P =0.003). The intera ctive model between early blight and heat wave was not 38 better than the additive model ( 2 =0.9, df=1, P =0.33) , suggesting that heat waves have similar e ffects on disease prevalence on control and early blight - inoculated plants . The heat wave treatment in n on - inoculated plants reduced natural occurring early blight by a non - significant 12% when compared to control treatments (t= - 3.016, df= 83.1, P =0.06). F or plants that were inoculated with early blight , adding a heat wave treatment reduced early blight by 1 3% (t= - 3.016, df=83.1, P =0.06) compared to control plants. The beetle treatment had no effect on early blight ( 2 =0.07, df=1, P=0.79). Figure 3.4 . Sum of percent dead potato foliage for all seven sampling dates combined. The circles represent plants with no CPB treatment and the triangles represent those that received 15 first instar Colorado potato beetles (blu e shapes = control and red = heat wav e (HW) treatment). 39 Potato yield P lants exposed to early blight treatments had lower yield ( 2 = 8.11, df = 1, P =0.004) regardless of heat wave or beetle treatments. The beetle treatments also had negative impacts on total tuber mass ( 2 = 5.15, df = 1, P =0.02). Heat wave treatments alone had no effect on total tuber mass ( 2 = 1.28, df = 1, P =0.26; Fig. 3.5. ). W e found a 20% reduction in total yield for plants that were exposed to early blight and beetles (t=3.53, d f=156, P =0.01), 17% re duction for heat wave and beetles (t=3.53, df=156, P =0.01), and 14% for early blight and heat wave (t=2.0, df=83.1, P =0.47). The exposure to all three stressors reduced yield by 21% (t=3.1, df=129.3, P =0.046) but did not have more det rimental effects than any of the dual stressor treatment combinations (t=0.56, df=83.1, P =0.99; t=2.3, df=83.1, P =0.3 0 ; t=2.7, df=81, P =0.1 1). We were interested in the effect heat waves would have on plants when combined with beetle damage so we used a su bset of the data to only include those treatments. We found that adding beetles to the heat wave treatment decreased total tuber yield by 2 4% ( 2 = 5.3, df = 1, P =0.02) compared to plants just experiencing a heat wave. 40 Figure 3.5 . Average total tuber ma ss produced per plant for control (blue points) and heat wave (HW) plants (red points) exposed to beetle (CPB), early blight inoculation treatments (Blight), and both beetle and early blight exposure (Blight+CPB). Circles denote pla nts without CPB treatmen ts and triangles represent plants exposed 15 first instar Colorado potato beetles . Points and error bars show means + - one standard error from 20 replicates per treatment. DISCUSSION In this study, I investigated the consequences of heat waves on the in teraction between three stre ssors of potato . First, I found that beetle larvae feeding on diseased plants , plants exposed to heat waves and plants experiencing both stressors simultaneously had higher larval mass indicating faster development rates . Second , heat waves slowed the progression of early 41 blight infecting and spreading among the leaves of potato plants. Finally, the effect of the heat wave, herbivory, and disease had no impact on potato tube r yield except when two stressors occurred simultaneousl y which caused up to 20% potato yield loss. Overall, heat waves influenced interactions between plants , herbivores, and diseases b y decreasing development rate of CPB a nd changing environmental condit ions to be less favorable for early blight and more favo rable for CPB . The results suggest that if there is a higher frequency of heat waves, then an additional CPB generation is likely to occur in the latitudes where these experiments were performed . The spread of early blight on infected plants was slowed b y the extreme heat conditions in our simulated heat wave experiment. In contrast to a pilot experiment that suggest ed more rapid disease spread in the presence of herbivores via movemen t of spores and wounding of leaves, I found the amount of disease was n ot greater for those plants that were inoculated with early blight and received herbivore damage. This experiment was conducted in the last month of the growing season and would likely have shown CPB to facilitate the spread of disease if the herbivory and disease treatment additions were timed so that beetle damage occu rred a week prior to disease inoculation as done in the pilot study. The larvae were only on the plant for 10 d , while herbivore pressure starts as soon as the plant emerges and continues fo r the duration of the plant s life cycle (approximately 90 - 100 d) which would likely increase the impact of herbivory on disease spread (Tsedaley, 2014). I nsect survival rate was not impacted by any of the treatments , one possible explanation of this is t hat the parent population was collected from the newly emerged summer generation of CPB ( o ffspring of the overwintering population) whi ch had been exposed to daily variation and the high heat in mid - late June and July, possibly allowing individuals to accl imate 42 physiologically to high heat and potentially passing that acclimation on to their offspring epigenetically . Our previous field experiments found significant reductions in survival of CPB larvae sourced from a parent colony kept in the lab at a consta Research suggests that offspring often inherit resista nce to stressors accumulated by their parents which might be the case here (Onstad 2014) . T his lack of mortality from heat treatments could also be due to behavioral changes whereby larvae were able to locate microclimates on the plant that allowed them to temper the effects of the extreme heat as observed in spiders a nd Tenebrionid beetles (Barton & Schmitz, 2009; Ward & Seely, 1996 ). The final in star achieved after 10 d of feeding on the plant was not affected by early blight but 5% more larvae that experienced the 5 - d heat wave treatment achieved the final 4 th instar. While this is not a drastic increase in development, those larvae in heat wave treatments had mass 44% greater than control sug gesting that those larvae in the heat wave treatment were near the end of their 4 th instar while those on control plants were either in their early stages of the 4 th instar or late stages of their 3 rd instar. Larval mass however, increased by 16% by feedin g on blighted plants but t he heat wave conditions seem to cancel out the benefits of feeding on blighted plants as larva e feeding on blighted plants in the heat wave treatment had similar larval mass as larvae feeding on healthy plants in the heat wave tre atment. This increase in larval mass and instar is evidence that heat waves may boost larval development cycles allowing for faster turnover of generations especially if there are more than one heat wave s in a season (Pulatov et al. 2016) . Potato yield is a key response variable from a production perspective ; it was most influenced by combination s of stressors but not by any stressor alone. Plants in the early blight and heat wave treatments had similar yields to that of the control, but when combining the stress of a heat wave with beetle damage there was around a 20% decrease in total yield. This suggests 43 that potato plants are capable of managing one stressor with minimal effect on yield, but when adding an additional stressor that requires a different d efense pathway, resources normally used for reproduction now have to be allocated to defend ag ainst the second stressor (Züst & Agrawal, 2017) . P lants that were exposed to all three stressors had similar yields to those that were exposed to any two stresso rs , suggesting that the effects of this set of stressors saturate at two and that adding a third stressor does not reduce potato growth any further . These yield reductions would likely be much worse if the plant was exposed to earlier inoculation of insect a nd disease pests, but the yield reduction seen here is evidence that heat waves during early reproductive stages of potato paired with simultaneous herbivore or disease pressure can have drastic impacts on yield . In this study, I provide d evidence that a 5 - d heat wave can have major consequences for the interactions of two pests and their crop host plant. Results indicated that heat waves can slow the infection rate of disease that thrive in humid conditions and accelerate herbivore development. The accel e rated development of the herbivore and slowed progression of disease will potentially allow for sufficient potato foliage for an extra generation of herbivores to feed on. The likelihood of an additional generation of herbivores may be higher if farmers p l ant later into the year taking advantage of the extended growing season. Heat waves caused insect and disease stressed plants to have reduced yields. Therefore , it is important to control insect pest infestations and disease infection prior to a forecaste d heat wave event. The yield losses are large enough that they may warrant applying a pesticide prior to reaching the recommended economic threshold. While yield is important these results show that multi - species interactions are heavily influenced by shor t duration, high intensity heat waves and justifies the attention of more ecological studies, especially since the frequency intensity and duration of heat waves continues 44 to rise (Meehl and Tebaldi 2004 ; Rahmstorf and Coumou 2011 ; Fischer and Knutti 2015 ; Perkins - Kirkpatrick and Gibson 2017 ) . Future climate warming scenarios on plant, herbivore, and disease interactions should include the consequences of brief but extreme stress events, such as heat waves. 45 APPENDIX 46 Table 3.1 Mean ± SEM temperatures (°C) for field heat wave experiments in Michigan in an experimental potato field in 2019. Heat waves were created using open top chambers and ceramic heaters. Day means are calculated during a three - hour interval from (14:00 - 16:00 h), night from (4:00 - 6:00 h) o n all five days of the heat wave treatment. Means were calculated from 5 Hobo data loggers in control and 5 data loggers in heat wave treatments. Also showing Relative humidity for the same time frames (no standard error only one logger per treatment). A ugust 5 - 10 2019 Control Heat wave Day Night Day Night Air temperature 32.1±0.5 17.2±0.1 38.1±0.6 21.1±0.2 Leaf surface 27.2±0.5 13.1±0.3 32.3±0.8 22.7±0.2 Soil surface 39.2±1.1 15.4±0.2 43.6±1.1 24.4±0.2 Soil sub - surface 25.6±0.3 22.9±0.1 28.7±0.2 26.6±0.1 % Rel. Humidity 48.6 93.6 41.3 76.9 Table 3.2 Differences between heat wave and control treatments in open top chambers for heat waves created in an experimental potato field in Michigan in 2019. Difference HW - Control Day Night Air temperature 6 3.9 Leaf temp 5.1 9.6 Soil surface 4.4 9 Soil sub - surface 3.1 3.7 Humidity - 7.3 - 16.7 47 CHAPTER 4: Conclusions and future directions The goal of this research was to improve our understand ing of the direct and indirec t effects of heat waves on insect, plant, and disease interactions including the role of heat wave timing and plant ontogeny, with the end goal of improving future crop management. In chapter 2 I found direct negative effects of heat waves on herbivore sur vival as well as indirect effects on survival for larvae feeding on a plant that was previously exposed to heat wave conditions one week earlier. The field experiment in 2018 showed the timing of heat wave s and herbivory is important for yield, with plants in early reproductive stages tuber initiation and e arly tuber bulking being the most affected by heat stress and herbivory. Insects feeding on 8 - week - old plants and 12 - week - old plants during a heat wave had the lowest survival rate of any treatments. A po tential explanation for the reduced survival on younger plants is defenses may have been concentrated by water loss during the heat wave reducing palatability for the herbivore. Larvae likely performed worse on older plants due to lower nutritional value in their foliage as resources are being directed towards tuber bulking and maturation. Although larval mass was consistently higher on heat wave treated plants, regardless of plant age, the only significant in crease in larval mass o ccurred in larvae feeding on those 8 - week - old plants. CPB are specialist herbivores that have evolved to cope with Solanaceous defensive compounds which likely explains the higher mass achieved by these larvae ( Groden & Casagrande, 1 986 ) . Marketable yield was most effected by the treatments compare d to total yield in our 2018 field experiment with smaller but still significant reductions in marketable yield in the 2019 experiment suggesting that heat wave and beetle treatments signif icantly alter tuber qua lity not just tuber production. This could result in major losses of revenue for farmers by having to sell 48 their crop to a different market and or lower price as it does not m eet quality standards. With heat waves in the Midwest occurring most commonly fro m early June to mid - August the risk of these sensitive life stages of potato being exposed to stressful temperatures is high, especially for later crop plantings (Anderson et al., 2011) . The third chapter investigated t he influence of heat waves on multi - species interactions. While no t reatment combinations influence d larval survival , I observed an increase in larval mass and development rate for those exposed to heat wave treatments. This suggests that heat waves may be beneficial to specialist insect he rbivores which could lead to outbreaks. The increased development rate may also increase the likelihood of there being an additional generation of herbivores prior to what is predicted in gradual warming models (Pulatov e t al. 2016) . This may be facilitate d by farmers taking advantage of the longer growing season and planting crops later in the year. We also found that heat waves decreased the infection by early blight likely because the temperatures and humidity achieved during the heat wave are outside th e optimal range for early blight sporulation (Chaerani & Voorrips, 2006). This delayed infection rate may prolong the availability of edible leaf tissue for herbivores furthering the possibility of an additional generatio n of insect pests . The impact of heat waves on the interaction of potato plants with an associated specialist herbivore and fungal pathogen showed minimal effects on yield from any single stressor. The evolutionary relationship between these or sim ilar organisms means that the plant likel y has well developed defense strategies for protection. While heat waves slightly increased total yield, the addition of beetle or early blight damage decreased yield by 24% and 21% respective to plants in the heat wave treatment. While yield is important for the bottom line, these results show that multi - species interactions are heavily influenced by short duration, high intensity heat waves and 49 justifies the attention of more ecological studies, especially since th e frequency, intensity and duration of he at waves continues to rise (Meehl and Tebaldi 2004; Rahmstorf and Coumou 2011; Fischer and Knutti 2015; Perkins - Kirkpatrick and Gibson 2017) . Scientists investigating future climate warming scenarios on either plant , herbivore, or disease interactions shou ld include the consequences of extreme stress events, such as heat waves. These future studies could be used to create a more intelligent pest prediction model that would inform growers of optimal times to control p est populations to protect yield. While t he duration of a heat wave may be short , the ecological and agronomic impact s are large and lasting. 50 APPENDI X 51 RECORD OF DEPOSITION OF VOUCHER SPECIMENS The specimens listed below have been deposited in the named museum as samples of those species or othe r taxa, which were used in this research. Voucher recognition labels bearing the voucher number have been attached or included in fluid preserved specimens. Voucher Number: 2020 - 0 8 Author : Joshua S Snook Title of thesis: EFFECTS OF HEAT WAVE TIMING O N PLANT, HERBIVORE, DISEASE INTERACTIONS Museum(s) where deposited: Albert J. Cook Arthropod Research Collection, Michigan State University (MSU) Specimens: Family Genus - Species Life Stage Quantity Preservation Chrysomelidae Leptinotarsa decem lineata adult 20 pinned 52 LITERATURE CITED 53 LITERATURE CITED Al - Whaibi, M. H. (2011). Plant heat - shock proteins: A mini review. In Journal of King Saud University - Science (Vol. 23, Issue 2, pp. 139 150). https:/ /doi.org/10.1016/j.jksus.2010.06.022 Alyokhin, A. (2009). Colorado Potato Beetle Management on Potatoes: Current Challenges and Future Prospects. In Fruit, Ve getable and Cereal Science and Biotechnology : Global Science Books Anderson, G. B., Brooke Anders on, G., & Bell, M. L. (2011). Heat Waves in the United States: Mortality Risk during Heat Waves and Effect Modification by Heat Wave Characteristics in 43 U.S . Communities. In Environmental Health Perspectives (Vol. 119, Issue 2, pp. 210 218). https://doi. org/10.1289/ehp.1002313 Bale, J. S., Masters, G. J., Hodkinson, I. D., Awmack, C., Martijn Bezemer, T., Brown, V. K., Butterfield, J., Buse, A., Coulson, J. C ., Farrar, J., Good, J. E. G., Harrington, R., Hartley, S., Hefin Jones, T., Lindroth, R. L., Malc olm C. Press, Symrnioudis, I., Watt, A. D., & Whittaker, J. B. (2002). Herbivory in global climate change research: direct effects of rising temperature on in sect herbivores. In Global Change Biology (Vol. 8, Issue 1, pp. 1 16). https://doi.org/10.1046/j.1 365 - 2486.2002.00451.x the impact of extreme heat and frost events on wheat crop production: A review. In Field Crops Research (Vol. 171, pp. 109 119). https: //doi.org/10.1016/j.fcr.2014.11.010 Barton, B. T., & Schmitz, O. J. (2009). Experimental warming transforms multiple predator effects in a grassland food web. Ecology Letters, 12(12), 1317 1325. Bisbis, M. B., Gruda, N., & Blanke, M. (2018). Potential impa cts of climate change on vegetable production and product quality A review. In Journal of Cleaner Production (Vol. 170, pp. 1602 1620). https://doi.org/10.1 016/j.jclepro.2017.09.224 Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R ., Stevens, M. H. H., & White, J. - S. S. (2009). Generalized linear mixed models: a practical guide for ecology and evolution. Trends in Ecology & Evolution, 2 4(3), 127 135. Brown, J. H., Gillooly, J. F., Allen, A. P., Savage, V. M., & West, G. B. (2004). T OWARD A METABOLIC THEORY OF ECOLOGY. In Ecology (Vol. 85, Issue 7, pp. 1771 1789). https://doi.org/10.1890/03 - 9000 Chaerani, R., & Voorrips, R. E. (2006). Tomato early blight (Alternaria solani): the pathogen, genetics, and breeding for resistance. In Jour nal of General Plant Pathology (Vol. 72, Issue 6, pp. 335 347). https://doi.org/10.1007/s10327 - 006 - 0299 - 3 54 Chen, J., Kitazumi, A., Alpuerto, J., Aly okhin, A., & de Los Reyes, B. (2016). Heat - induced mortality and expression of heat shock proteins in Colorad o potato beetles treated with imidacloprid. Insect Science, 23(4), 548 554. Chiu, M. - C., Kuo, J. - J., & Kuo, M. - H. (2015). Life stage - dependent effe cts of experimental heat waves on an insect herbivore. In Ecological Entomology (Vol. 40, Issue 2, pp. 175 181). https://doi.org/10.1111/een.12173 Douglas, D. R. (1972). The effect of light and temperature on the sporulation of different isolates of Altern aria solani. In Canadian Journal of Botany (Vol. 50, Issue 3, pp. 629 634). https://doi.org/10.1139/b72 - 077 Dowd, W. W., King, F. A., & Denny, M. W. (2015). Thermal variation, thermal extremes and the physiological performance of individuals. The Journal o f Experimental Biology, 218(Pt 12), 1956 1967. Dyer, L. A., Richards, L. A., Short, S. A., & Dodson, C. D. (2013). Effects of CO2 and temperature on tritrophic interactions. PloS One, 8(4), e62528. Dyer, L., & Letourneau, D. (2013). Can Climate Change Trig ger Massive Diversity Cascades in Terrestrial Ecosystems? In Diversity (Vol. 5, Issue 3, pp. 479 504). https://doi.org/10.3390/d5030479 Ferrise, R., Moriondo, M., Trombi, G., Miglietta, F., & Bindi, M. (2013). Climate Change Impacts on Typical Mediterranea n Crops and Evaluation of Adaptation Strategies to Cope With. In Advances in Global Change Research (pp. 49 70). https://doi.org/10.1007/978 - 94 - 007 - 5772 - 1_4 Ferro, D. N., Logan, J. A., Voss, R. H., & Elkinton, J. S. (1985). Colorado Potato Beetle (Coleopte ra: Chrysomelidae) Temperature - dependent Growth and Feeding Rates. In Environmental Entomology (Vol. 14, Issue 3, pp. 343 348). https://doi.org/10.1093/ee/14.3.343 Fischer, E. M., & Knutti, R. (2015). Anthropogenic contribution to global occurrence of heav y - precipitation and high - temperature extremes. In Nature Climate Change (Vol. 5, Issue 6, pp. 560 564). https://doi.org/10.1038/nclimate2617 Groden, E., & Casagrande, R. A. (1986). Population Dynamics of the Colorado Potato Beetle, Leptinotarsa decemlineat a (Coleoptera: Chrysomelidae), on Solanum berthaultii. In Journal of Economic Entomology (Vol. 79, Issue 1, pp. 91 97). https://doi.org/10.1093/jee/79.1.91 Harrington, R., Woiwod, I., I., & Sparks, T. (1999). Climate change and trophic interactions. Trends in Ecology & Evolution, 14(4), 146 150. Intergovernmental Panel on Climate Change. (2014). Climate Change 2014 Impacts, Adaptation and Vulnerability: Part B: Regional Aspects: Volume 2, Regional Aspects: 55 Working Group II Contribution to the IPCC Fifth A ssessment Report. Cambridge University Press. Leisner, C. P., Wood, J. C., Vaillancourt, B., Tang, Y., Douches, D. S., Robin Buell, C., & Winkler, J. A. (2018). Impact of choice of future climate change projection on growth chamber experimental outcomes: a preliminary study in potato. International Journal of Biometeorology, 62(4), 669 679. Lesk, C., Rowhani, P., & Ramankutty, N. (2016). Influence of extreme weather disasters on global crop production. Nature, 529(7584), 84 87. Levy, D., & Veilleux, R. E. ( 2007). Adaptation of potato to high temperatures and salinity - a review. In American Journal of Potato Research (Vol. 84, Issue 6, pp. 487 506). https://doi.org/10.1007/bf02987885 F. (2012). Extreme climatic events and vegetation: the role of stabilizing processes. In Global Change Biology (Vol. 18, Issue 3, pp. 797 805). https://doi.org/10.1111/j.1365 - 248 6.2011.02624.x Logan, P. A., Casagrande, R. A., Faubert, H. H., & Drummond, F . A. (1985). Temperature - dependent Development and Feeding of Immature Colorado Potato Beetles, Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae). In Environmental Entom ology (Vol. 14, Issue 3, pp. 275 283). https://doi.org/10.1093/ee/14.3.275 Me ehl, G. A., & Tebaldi, C. (2004). More intense, more frequent, and longer lasting heat waves in the 21st century. Science, 305(5686), 994 997. Myers, S. S., Smith, M. R., Guth, S. , Golden, C. D., Vaitla, B., Mueller, N. D., Dangour, A. D., & Huybers, P. (2 017). Climate Change and Global Food Systems: Potential Impacts on Food Security and Undernutrition. Annual Review of Public Health, 38, 259 277. Neven, L. G. (2000). Physiologica l responses of insects to heat. In Postharvest Biology and Technology (Vol. 2 1, Issue 1, pp. 103 111). https://doi.org/10.1016/s0925 - 5214(00)00169 - 1 Onstad, D. W. (2014). IPM and Insect Resistance Management. In Insect Resistance Management (pp. 515 532). https://doi.org/10.1016/b978 - 0 - 12 - 396955 - 2.00016 - 3 Parmesan, C., & Yohe, G. ( 2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature, 421(6918), 37 42. Perkins - Kirkpatrick, S. E., & Gibson, P. B. (2017). Changes in r egional heatwave characteristics as a function of increasing global temperatu re. Scientific Reports, 7(1), 12256. Texas Medicine, 116(5), 18 23. 56 Pulatov, B., Jö nsson, A. M., Wilcke, R. A. I., Linderson , M. - L., Hall, K., & Bärring, L. (2016). Evaluation of the phenological synchrony between potato crop and Colorado potato beetle under future climate in Europe. In Agriculture, Ecosystems & Environment (Vol. 224, pp . 39 49). https://doi.org/10.1016/j.agee. 2016.03.027 Rahmstorf, S., & Coumou, D. (2011). Increase of extreme events in a warming world. Proceedings of the National Academy of Sciences of the United States of America, 108(44), 17905 17909. R Core Team (2019 ). R: A language and environment for stat istical computing. R Foundation for Statistical Computing, Vienna, Austria. Version 3.5.3. https://www.R - project.org/ Robinson, P. J. (2001). On the Definition of a Heat Wave. In Journal of Applied Meteorology (Vol. 40, Issue 4, pp. 762 775). https://doi.o rg/2.0.co;2">10.1175/1520 - 0450(2001)040<0762:otdoah>2.0.co;2 Rosenzweig, N., G. Olaya, Z. K. Atallah, S. Cleere, C. Stanger, and W. R. Stevenson. 2008. Fungicide in Alternaria Solani in Wiscon 60. Sentis, A., Hemptinne, J. - L., & Brodeur, J. (2013). Effects of simulated heat waves on an experimental plant - herbivore - predator food chain. Global Change Biology, 19(3), 833 842. Siebers, M. H., Slattery, R. A., Yendrek, C. R., Locke, A. M., Drag, D., Ainsworth, E. A., Bernacchi, C. J., & Ort, D. R. (2017). Simulated heat waves during maize reproductive stages alter reproductive growth but have no lasting effect when applied during vegetative stages. In Agriculture, Ecosy stems & Environment (Vol. 240, pp. 162 170). https://doi.org/10.1016/j.agee.2016.11.008 Siebers, M. H., Yendrek, C. R., Drag, D., Locke, A. M., Rios Acosta, L., Leakey, A. D. B., Ainsworth, E. A., Bernacchi, C. J., & Ort, D. R. (2015). Heat waves imposed d uring early pod development in soybean (Glycine max) cause significant yield loss despite a rapid recovery from oxidative stress. Global Change Biology, 21(8), 3114 3125. Smith, M. D. (2011). An ecological perspecti ve on extreme climatic events: a syntheti c definition and framework to guide future research. In Journal of Ecology (Vol. 99, Issue 3, pp. 656 663). https://doi.org/10.1111/j.1365 - 2745.2011.01798.x Szczepaniec, A., & Finke, D. (2019). Plant - Vector - Pathogen Interactions in the Context of Drought S tress. In Frontiers in Ecology and Evolution (Vol. 7). https://doi.org/10.3389/fevo.2019.00262 Thomma, B. P. H. J. (2003). Alternaria spp.: from general saprophyte to specific parasite. Molecular Plant Pathology, 4( 4), 225 236. 57 Thompson, R. M., Beardall, J ., Beringer, J., Grace, M., & Sardina, P. (2013). Means and extremes: building variability into community - level climate change experiments. Ecology Letters, 16(6), 799 806. Tsedaley, B. (2014). Review On Early Bligh t (Alternaria spp.) of Potato Disease and its Management Options. In Journal of Biology, Agriculture and Healthcare (Vol. 4, No.27, pp. 191 199) von Grebmer, K., von Grebmer Maximo Torero Tolulope Olofinbiyi Heidi Fritschel Doris Wiesmann Yisehac Yohannes, K., Torero, M., Olofinbiyi, T., Fritsche l, H., Wiesmann, D., Yohannes, Y., Schofield, L., & von Oppeln, C. (2011). 2011 Global Hunger Index The challenge of hunger: Taming price spikes and excessive food price volatility. Intl Food Policy Res Inst. Ward, D., & Seely, M. K. (1996). Behavioral The rmoregulation of Six Namib Desert Tenebrionid Beetle Species (Coleoptera). In Annals of the Entomological Society of America (Vol. 89, Issue 3, pp. 442 451). https://doi.org/10.1093/aesa/89.3.442 Ward, D., & Seely, M. K. (1996). ADAPTATION AND CONSTRAINT I N THE EVOLUTION OF THE PHYSIOLOGY AND BEHAVIOR OF THE NAMIB DESERT TENEBRIONID BEETLE GENUS ONYMACRIS. Evolution; International Journal of Organic Evolution, 50(3), 1231 1240. White, T. J., T. Bruns, S. Lee, and J. S PCR Protocols. https://doi.org/10.1016/b978 - 0 - 12 - 372180 - 8.50042 - 1. Yang, L. H., & Rudolf, V. H. W. (2010). Phenology, ontogeny and the effects of climate c hange on the timing of species interactio ns. In Ecology Letters (Vol. 13, Issue 1, pp. 1 10). https://doi.org/10.1111/j.1461 - 0248.2009.01402.x Zhang, W., Rudolf, V. H. W., & Ma, C. - S. (2015). Stage - specific heat effects: timing and duration of heat waves a lter demographic rates of a global insect pest. In Oecologia (Vol. 179, Issue 4, pp. 947 957). https://doi.org/10.1007/s00442 - 015 - 3409 - 0 Züst, T., & Agrawal, A. A. (2017). Trade - Offs Between Plant Growth and Defense Against Insect Herbivory: An Emerging Mechanistic Synthesis. Annual Review of Pla nt Biology, 68, 513 534.