Experimental design: To understand how the mode of exposure of sublethal levels of microcystin-LR (MCLR) impacts a keystone and ecological engineer species, environmentally relevant concentrations of MCLR will be delivered to crayfish via water or soil as well as a control without MCLR. The effect of MCLR mode of delivery will be assessed using crayfish foraging and bioturbation behavior. To mimic an acute exposure, crayfish will be dosed for 6 days with a sublethal concentration of MCLR (0.5 μg / L) in small-scale dosing tanks. After the dosing period, crayfish will be moved to an experimental arena filled with flowing clean water and soil to monitor their foraging and bioturbation behavior through a consumption assay and a behavioral assay during an overnight period. There will be a total of 10 trials. Collection and Housing of Organisms: Crayfish Faxonius virilis will be collected from Maple Bay in Burt Lake (45.4802° N, 84.6974° W) using hand nets. After collection, crayfish will be transported to the University of Michigan Stream Research Facility for housing. Crayfish will be placed in a flow through cattle trough (200 x 60 x 60 cm, l x w x h) which will be fed from the east branch of the Maple River. Unfiltered water will flow into the tank from a PVC delivery pipe and exit the tank via a standpipe which will keep the water depth at approximately 60 cm. PVC pipes will be placed in the cattle through to be used as crayfish shelters (Austin & Moore, 2022). The cattle trough will be placed outside and exposed to natural light:dark cycles as well as natural temperature fluctuations. Only form II females with intact appendages will be used within the trials. Crayfish will feed on the detritus that arrives via the flow of water. Carapace length (IOL) and the right chelae of each crayfish will be measured to the nearest 0.05 cm before use in a trial. Microcystin-LR Exposure: Microcystin will be purchased from Musechem. Dosing regimens will occur in glass aquaria (40.5 x 25.5 x 20.5 cm, l x w x h) which will be filled with 15 L of unfiltered water and 4.0 kg of sand taken from the Maple River. For water-dosed studies, MCLR will be added to the water to produce a final concentration of 0.5 μg/L. For sediment-dosed studies, MCLR will be mixed with the sediment and will be allowed to sit for 24 hours for a final concentration of 0.5 ug/L. For control studies, no MCLR will be added to the system. Crayfish will remain in the exposure tanks for 6 days. Exposure tanks will be dosed once before each trial. Each tank will be aerated with an air bubbler. Experimental arena: The experimental arena will consist of concrete cinder blocks covered with plastic lining and will be used to make six flow-through stream mesocosms (1.5 x 0.5 x 0.28 m, l x w x h) (Austin & Moore, 2022; Moore & Grills, 1999; Wolf et al., 2004). Three 208 L plastic drums will serve as constant head tanks for the six mesocosms. These plastic drums will be filled with water from the Maple River. Each plastic drum will feed two hoses into six mesocosms with water from a 1 cm diameter garden hose with a flow rate of 0.086 ± 0.003 L/s [mean ± SEM]. Nylon stockings will cover the ends of the supply pipes to filter macroinvertebrates and organic matter (Austin & Moore, 2022). The bottom of the experimental mesocosms will be lined with approximately 191.8 kg of sand collected from the Maple River. Red light bulbs will be used to illuminate the system. A wooden frame will hold a Swann Pro-Series DVR4-3000TM TruBlueTM camera 1.5 m above the surface of the water to record crayfish behavior (Wood & Moore, 2019). Food Production: Fish gelatin will be prepared by blending 0.3 L boiled water, 20 g sardines, and 12.5 g Knox unflavored gelatin mix. After blending, the gelatin will be placed into plastic caps, covered with plastic wrap, and refrigerated overnight. Each cap will be filled with approximately 5 g of food, and fish gelatin will not be used longer than 72 h after preparation (Austin & Moore, 2022; Edwards et al., 2018; Jackson & Moore, 2019). Experimental protocol: At the beginning of each trial, one crayfish will be placed into each exposure tank at 2345. After six days in the exposure tanks, crayfish will be relocated into the experimental mesocosms at 2345, allowing the crayfish 15 minutes to acclimate to their new environment. Crayfish will be removed from the experimental mesocosms at approximately 0800 the next morning. Fish gelatin will be placed in the experimental mesocosm at 0000 and will be promptly removed the following morning at 0800. After removal, the fish gelatin will be gently spin-dried in a salad spinner and weighed (Austin & Moore, 2022). From 0000-0400 the cameras will record crayfish behavior (Wood & Moore, 2019). Data Collection: Crayfish consumption will be calculated by subtracting the initial and final weights of the fish gelatin. The amount consumed will be calculated by using the following formula: Gelatin consumption = (Gf- Gi)/Gi x 100, where Gf equals the final weight of the fish gelatin and where Gi equals the initial weight of the fish gelatin (Austin & Moore, 2022). Crayfish behavior from the video-recordings will be measured by watching the video-recordings and categorizing behaviors specified in Table 1 by the second. Table 1: Ethogram describing specified crayfish behaviors pertinent to foraging and bioturbation. Foraging - The crayfish is within 10 cm of the food source. Digging - The crayfish is using its chelae, its two front walking legs, or its entire body to push sand, or carry and displace sand around the enclosure. Locomotion - The crayfish is changing the location of its body at 1 cm/s or greater. Data Analysis: Consumption of fish gelatin will be assessed using a nonlinear mixed model by using the “nlmer()” function on the lme4 package in R (Bates et al., 2023). The fish gelatin consumption model was constructed using one categorical fixed factor (mode of MCLR dosing) and two random effects (experimental mesocosm and trial) with trial nested within mesocosm. Outputs will be obtained by performing a one-way analysis of variants using the “ANOVA” function of the “car” package (Fox & Weisburg, 2019). To extract behavioral data from the assays, an observer blind to the treatment will quantify the position of the crayfish every 15 seconds. From these x, y coordinates, time crayfish spent engaged in the behaviors outlined in the ethogram (Table 1) will be produced. Time spent performing behaviors will be assessed using a nonlinear mixed model by using the “nlmer()” function on the lme4 package in R (Bates et al., 2023). Each video-taped behavioral category analyzed was constructed with full interactions using one categorical fixed factor (mode of MCLR dosing) and two random effects (experimental mesocosm and trial) with trial nested within mesocosm. Outputs will be obtained by performing a one-way analysis of variants using the “ANOVA” function of the “car” package for each of the four dependent variables (Fox & Weisburg, 2019).