Coffee grown in the United States represents less than 1% of the world's supply, and most of it comes from the state of Hawai'i. Kona coffee, grown on the western side of the island of Hawai'i, is the most recognized and the highest value Hawaiian coffee. The majority of this coffee is sun-dried after harvest and washing. Sun-dried coffee should reach 12-13% moisture within 4 to 6 days. Sun-drying will reduce both the moisture content and the water activity (aw). Reducing aw below 0.75, especially in the first week of drying, is important for preventing or limiting mold growth. The purpose of this study was to 1) compare drying rates of Kona coffee bean batches using aw and moisture content % measurements, 2) evaluate factors affecting the drying time of sun-dried Kona coffee, and 3) provide recommendations for post-harvest processing of sun-dried Kona coffee to optimize quality and safety. Ten farms in the Kona coffee region of Hawai'i were visited in the fall of 2017 to record data on the drying rate of coffee bean batches and to record observations on the post-harvest handling and storage of coffee beans and the environmental conditions that may affect the quality and microbial contamination of drying coffee. The coffee drying surfaces, physical enclosures, fan use, and elevation varied among farms. Daily measurements of coffee moisture level, water activity, depth of bean layer and temperature were recorded along with air temperature, relative humidity and cloud cover during drying for 30 batches. Most sun-dried batches reached 13% moisture in 6 to 10 days. Initial moisture content (31.6 4.3%), and drying yard characteristics varied greatly among farms. Coffee batches reached 0.75 aw within 6 days on average, but some batches required more than 10 days. Moisture content and aw measurements were weakly correlated and water activity level increased at times during drying for some batches. Allowing airflow around drying beans and maintaining a bean layer depth of less than 5 cm appeared to improve drying rates. Drying coffee parchments within 7 days post-harvest can inhibit growth of molds that may impact quality or molds that could produce mycotoxin. Controlling the drying conditions, including raking the layered beans, and monitoring moisture content can accelerate, or improve the consistency of, batch drying time. / Doctor of Philosophy / Coffee grown in the United States represents less than 1% of the world’s supply, and most of it comes from the state of Hawai’i. Kona coffee, grown on the western side of the island of Hawai’i, is the most recognized and the highest value Hawaiian coffee. The majority of this coffee is sun-dried after harvest and washing. Sun-dried coffee should reach 12-13% moisture within 4 to 6 days. Sun-drying will reduce both the moisture content and the water activity (aw*). (*Water activity is the measurement of water vapor pressure generated by the free or non-chemically bound water in foods and other products. Water activity (range of 0 to 1) is an important indicator for the shelf life of foods and the occurrence and growth of microorganisms). Reducing aw below 0.75, especially in the first week of drying, is important for preventing or limiting mold growth. The purpose of this study was to 1) compare drying rates of Kona coffee bean batches using aw and moisture content % measurements, 2) evaluate factors affecting the drying time of sun-dried Kona coffee, and 3) provide recommendations for post-harvest processing of sun-dried Kona coffee to optimize quality and safety. Ten farms in the Kona coffee region of Hawai’i were visited in the fall of 2017 to record data on the drying rate of coffee bean batches and to record observations on the post-harvest handling and storage of coffee beans and the environmental conditions that may affect the quality and microbial contamination of drying coffee. The coffee drying surfaces, physical enclosures, fan use, and elevation varied among farms. Daily measurements of coffee moisture level, water activity, depth of bean layer and temperature were recorded along with air temperature, relative humidity and cloud cover during drying for 30 batches. Most sun-dried batches reached 13% moisture in 6 to 10 days. Initial moisture content (27 to 36 %), and drying yard characteristics varied greatly among farms. Coffee batches reached 0.75 aw within 6 days on average, but some batches required more than 10 days. Moisture content and aw measurements were weakly correlated and water activity level increased at times during drying for some batches. Allowing airflow around drying beans and maintaining a bean layer depth of less than 5 cm appeared to improve drying rates. Drying coffee parchments within 7 days post-harvest can inhibit growth of molds that may impact quality or molds that could produce mycotoxin. Controlling the drying conditions, including raking the layered beans, and monitoring moisture content can accelerate, or improve the consistency of, batch drying time.

Interest in wood-based bio-energy systems in the United States is increasing and may play a part in future renewable energy initiatives (Dincer 2000). Family forests have potential to play an important role in supplying wood biomass for energy production. However, access depends mostly on the management intentions among family forest owners. Enhanced biomass markets in regions where family forest ownership dominates could increase productivity by reinvigorating the low-value merchandizing required to accomplish silvicultural objectives. Given diverse owner objectives and forest types on family forests, estimates of biomass availability must include both biophysical and social aspects of procurable feedstock. This thesis chronicles a biosocial case study that estimates potential biomass supply from 51 family forests in Virginia and North Carolina. The study occurred within a woodshed centered on the future site of an impending ethanol plant in Mecklenburg County, Virginia. A survey instrument using the theory of planned behavior was used to measure ownership characteristics and intention to harvest. Forest attributes were collected during property visits to estimate potential yields resulting from silvicultural simulations. Results reveal that forest cover-type and tree size significantly affect owner intentions to harvest and owner attitudes toward harvesting partially mediate this relationship. Outputs from silvicultural simulations correspond with those made using Forest Inventory and Analysis data within the study region. Disproportionality was examined by coupling social and biological drivers of sustainable wood biomass availability. Implications of the research include refined estimates of potential supply and demonstrating a multi-scalar, mixed-method approach for assessing wood biomass availability. / Master of Science

Research documenting the functional changes of wetlands in response to harvest disturbance has often been limited to initial-response time frames of one or two years. A cypress (Taxodium distichum L. Rich.)-water tupelo ~ aQuatica L.) swamp located in the Mobile-T ensaw River Delta of Southwestern Alabama was harvested in the fall of 1986. Harvest treatments included: c1earcutting the entire study area with helicopter log removal, or with an added simulated rubber-tired skidder transport, or with herbicide control of all regenerating vegetation. An adjacent nondisturbed stand served as a reference area. Remeasurement of the area was initiated seven years after harvest. The helicopter logged and skidder simulation treatments have produced stands of comparable diameter, height, density, and basal area. ~ aqyatica was dominant in the skidder simulation treatment~ ~ tica, ~ nigra, and Fraxinus caroliniana dominated the helicopter treatment. Overstory biomass was less in the helicopter logged areas (20,981 kg/ha) than in the skidder simulation treatment (30,533 kglha). Lowerstory trees in both treatments were of comparable diameter and height, however, the helicopter treatment contained a greater density, basal area, and biomass ofFraxinus caroliniana. The rutting and mounding associated with the skidder simulation treatment favored a greater diversity of both diversity of both lowerstory and herbaceous species. Groundflora biomass was greatest in the herbicide controlled treatment (13,127 kg/ha), comparable in both the helicopter logged and skidder simulation treatments (5128 kg/ha; 5648 kglha, respectively), and least in the reference area (1495 kglha). Sediment accumulation reflected these differences in groundflora biomass; 104 mm in the herbicide controlled treatment, 79 mm and 63 mm in the helicopter logged and skidder simulation treatments, respectively, and 37 mm in the reference area. Three gradients were identified in the study areas via the tests of Latin squares, rows, and columns which actually corresponded to gradients of hydrology and! or sediment deposition. Square two was wetter than squares one or three because square two had fewer residual pullboat runs, which facilitate greater water movement. The rows are indicative of the distance from the main river channel; rows nearer the river are slightly higher in elevation, less inundated, and favor the less flood tolerant species. The columns reflect the sediment trapping patterns of the vegetation and the subsequent effects of site hydrology and nutritional status. / Master of Science

Solitary Vireos (Vireo solitarius alticola) were studied on an extensive mature forest landscape in 1994 and on three sites harvested using the group selection method in 1995. Pairing success was 88% on the unharvested site in 1994, however it was lower on the harvested sites in 1995 (53.8%). Low reproductive success was observed in both years (12% and 22% respectively). In 1995, the amount of opening within a 200 meter radius of the first observation point of a male was positively related to pairing success (p<0.05). This suggests that females may be cueing in on the increased number of shrubs in and around openings. / Master of Science

Amphibians, because of their semi-permeable skin, sensitivity to changing microclimates, and important role in ecosystems, are often viewed as indicators of ecosystem health. They make excellent organisms for studies on the effects of silvicultural practices. My goal was to provide recommendations for forest management in the southern Appalachians so that harvesting operations are compatible with maintaining healthy populations of forest amphibians. I tested previously created habitat models that determined the most important habitat characteristics for salamanders. I counted salamanders in 240 10 x 10 m plots located in the MeadWestvaco Wildlife and Ecosystem Research Forest in north-central West Virginia. We also collected a variety of habitat data in these plots to predict salamander abundance with previously created models. These simple linear regression analyses of predicted versus observed values suggest for most models (7 out of 9) a weak relationship between predicted and observed values (R2 from 0.0033 to 0.2869, p from < 0.0001 to 0.7490). However, one of the models showed characteristics suggesting that it predicted new data as well or better than the original data, and therefore was the most accurate at predicting salamander abundance, and could be used for management purposes, although there was still much unexplained variation. This model included the variables woody stems (< 7.5 cm DBH), available rock, riparian status (i.e., within 15 m of a stream), percent overstory canopy cover, and available highly decomposed woody debris (decomposition classes 3 to 5). All of these relationships were positive except for woody stems, suggesting that in order to maintain healthy populations of salamanders, we should protect areas next to streams, with high amounts of rock, decomposed woody debris, overstory canopy cover, and few woody stems. I also examined the immediate effects of clearcuts on salamander movement and population estimates. I batch marked salamanders in plots at the edges of a clearcut, and in a control plot. Using the Schnabel estimator, I estimated population sizes in each plot. I then compared population estimates pre- and post-harvest on the interior (harvested) and exterior (unharvested) sides of the plots, taking into account the control plot. I also examined adult-juvenile ratios and movements from one side of the plot to the other. I found no significant changes (p > 0.05) following harvest in any of these measures, suggesting that salamanders do not move out of the harvested area post-harvest, at least over the short term (10 months of this study). This suggests that a longer period of time (> 1 year) is required to observe the population declines detected in most studies. / Master of Science

Insect damage in stored maize is one of the major post-harvest losses occurring in developing countries, especially in sub-Saharan Africa. To reduce economic losses, separation of infested grains from clean ones is critical and requires detection of infestation in stored grains. This study aimed at developing a rapid non-destructive detection system with three goals; - the first goal was to investigate the acoustic behavior of internally feeding S. zeamais (Coleoptera: Curculionidae) in maize. The pests were monitored for peak activity and signal amplitudes. The study was performed in the morning, afternoon and night times for three consecutive days with an ultrasonic probe attached to acoustic emission detector. The average number of peaks above threshold in the morning, afternoon and night was significantly different and were 60, 2 and 31 counts/s, respectively (P-value < 0.01). The average maximum amplitude was also different: 2.5, 1 and 1.8V for morning, afternoon and night sessions, respectively. The signal frequencies ranged between 1 and 15 kHz with a peak around 7 kHz. The second goal was to design and develop an inexpensive acoustic device for the detection of S. zeamais in the stored maize. This device included a microphone, signal conditioning circuit and a microcontroller. The third goal was to test the prototype in both clean and infested maize. The device could be manufactured for $55 or less. The device has a noise level below 0.2V in clean maize, infestation amplitude up to 1V and about 93.3% correct detection performance in infected maize. / Master of Science

Herbicide resistant weeds are driving implementation of integrated weed management (IWM). A new tactic to manage weeds is harvest weed seed control (HWSC), which targets weed seeds retained on the plant at crop harvest and either destroys, removes, or concentrates them. Research is limited on the effectiveness of HWSC in US cropping systems. For HWSC to be effective it is important to know when and how many seed are shed from a weed species in relation to crop harvest. Research was conducted to quantify when weed seed are shattered from 6 economically important weed species, four broadleaf (redroot pigweed, common ragweed, common lambsquarters, and common cocklebur) and two grass species (large crabgrass and giant foxtail). Results indicate that among summer annuals, broadleaf species retain larger proportions of their seed compared to grass species at the first opportunity for soybean harvest. As harvest was delayed, more seeds shattered from all species evaluated, indicating timely harvest is critical to maximizing HWSC effectiveness. Studies were conducted on grower fields in Virginia to evaluate the effectiveness of HWSC (field residue and weed seed removal). Results indicate that HWSC can significantly reduce populations of Italian ryegrass in wheat and common ragweed in soybean in the next growing season, but reductions were not observed for Palmer amaranth in soybean. Investigating IWM system for common ragweed control in soybean, HWSC was found to be less effective than soybean planting date (i.e. double cropping after wheat) at reducing common ragweed populations. However, the effectiveness of HWSC varied by location. If HWSC adoption were to become widespread, weeds could adapt by shedding seed earlier in the season. Research was conducted by growing Palmer amaranth populations from across the eastern US in a common garden. Currently there are differences in flowering time and seed shatter among Palmer amaranth populations based on the location of the maternal population, indicating potential for adaptation. This research demonstrates that HWSC is a viable option for weed management in US cropping systems but needs to be stewarded like any other weed management tool. / Doctor of Philosophy / Herbicide resistance in weeds is a growing problem in the US and around the world. Alternative methods of weed control must be adopted to maintain crop yields in the presence of herbicide-resistant weeds. Researchers and extension specialists strongly advise growers to adopt an integrated weed management (IWM) approach. Integrated weed management involves implementing multiple weed control tactics during a growing season. By using multiple methods of weed control within a given season the chances of weeds becoming resistant or adapting to any single tactic is reduced. Harvest weed seed control (HWSC) is a new tactic developed in Australia in response to herbicide resistance. HWSC targets weed seeds retained on the plant at crop harvest. In a normal crop harvest, the combine removes the grain and spreads crop residues (leaves, stalks, and other plant parts), including weed seeds, back across the field. When HWSC is implemented, weed seeds are destroyed (narrow windrow burning, cage mills) or concentrated and potentially removed from the field (chaff carts, direct bale, chaff lining). Thus, HWSC limits the number of weed seeds returned to the soil seed bank. There is limited research on HWSC and its integration with other tactics, in US cropping systems. For HWSC to be effective it is necessary for weed seeds to be retained on the mother plant in sufficient quantities at crop harvest. Research was conducted in Virginia to determine when weed seeds are shattered during the soybean growing season for 6 economically important weed species, four broadleaf (redroot pigweed, common ragweed, common lambsquarters, and common cocklebur) and two grass species (large crabgrass and giant foxtail). The broadleaf species retained >85% of their seed until the first opportunity for soybean harvest (mid-October). In the grass species, more seed shattered prior to soybean harvest with 50% of large crabgrass and 74% of giant foxtail seed being retained at the first opportunity for soybean harvest. When harvest was delayed seed continued to shatter and less was captured using HWSC. This research indicates broadleaf species are more suitable candidates for HWSC than grass species, among summer annuals. Further research on the ability of seed to germinate in relation to when seeds were shed was conducted on redroot pigweed, common ragweed and common lambsquarters. Results indicate that there are variable effects on germination of these species depending on when they were shed. HWSC was implemented on grower fields to assess the impact on weed populations of 3 weed species (Italian ryegrass, common ragweed, and Palmer amaranth). These experiments compared conventional harvest and HWSC (field residue and weed seed removal) when all other management strategies were the same within that field. Italian ryegrass tiller density in wheat varied by location but was reduced up to 69% in the spring following implementation of HWSC. By wheat harvest, HWSC reduced Italian ryegrass seed head density 67% at one location compared to conventional harvest. In soybean, common ragweed densities were reduced by 22 and 26% prior to field preparation and postemergence herbicide applications, respectively, in the HWSC plots compared to the conventional harvest plots. No differences were observed in common ragweed density by soybean harvest. No differences were observed with Palmer amaranth densities at any point during the soybean growing season. This research show that HWSC can reduce weed populations but is variable and additional research is still needed. IWM experiments were established across Virginia to compare soybean planting date (full season or double cropped), + cover crop (cereal rye/wheat or no cover), and + HWSC (field residue removal) to evaluate the best management strategy for common ragweed in soybean. Across all locations, double cropping soybean behind wheat had the greatest impact on common ragweed densities at the end of the first season. The impact of double cropping soybeans on common ragweed population is due to the emergence pattern of common ragweed; majority of common ragweed emerges prior to planting double cropped soybean (mid-June to early-July). HWSC was variable and only reduced common ragweed density at one of three locations. Widespread adoption of HWSC could place a selection pressure on weeds to shatter seed earlier in the season. A common garden experiment was conducted in Blacksburg, VA to assess Palmer amaranth populations collected from central Florida to southern Pennsylvania for differences in flowering time, time to seed shatter, and other phenotypic traits. Results indicate that latitude of the maternal population influences time to first flower with a 0.53 d reduction in flowering time for every degree north in latitude the maternal population was collected from. The strongest predictor of Palmer amaranth flowering time was emergence date/daylength. For every day emergence was delayed the time to first flower was reduced by 0.31 and 0.24 d for female and male plants, respectively. Time from emergence or first flower to first seed shatter was reduced by 0.48 or 0.17 d, respectively, for each day emergence was delayed. These results indicate that differences exist currently among Palmer amaranth populations and the selection pressure of HWSC could push these populations to flower and shatter seed early.

Raw vegetables can sometimes be the source of outbreaks of human illness; however the potential for fresh vegetables to serve as a vehicle for antibiotic -resistant bacteria is poorly understood. Antibiotics and antibiotic-resistant bacteria have been shown to persist in manure of animals administered antibiotics, and in compost generated from this manure, where there is the potential for their transfer to produce. The purpose of this study was to determine the survival of antibiotic-resistant bacteria on raw, peeled, carrots after washing with commonly used chemical sanitizers. Multi-drug resistant E. coli O157:H7 and Pseudomonas aeruginosa were inoculated into a compost slurry of composted manure from dairy cattle, with and without prior administration of antibiotics, and used to inoculate carrot surfaces prior to the washing studies. This approach provided defined model antibiotic-resistant pathogens present within a background microbial community simulating potential carry over from manure-derived fertilizer. Carrots (n=3, 25g) were air-dried and stored at 4 °C until washing with tap water, XY-12 (sodium hypochlorite, 50 ppm free chlorine) or Tsunami 100 (peroxyacetic acid/hydrogen peroxide, 40 ppm free paracetic acid), according to manufacturer's directions. A second batch of carrots representing each inoculation x wash condition (n=3) were individually packaged for storage at 2 °C for 1,7, and 14 days, or 10 °C for 7 days and enumerated on those day intervals to recover bacteria from the surfaces of washed carrots. The resulting previously washed and stored carrots were subject to serial dilution and plated onto corresponding agar to enumerate total aerobic bacteria (R2A), aerobic bacteria tolerant or resistant to antibiotics (antibiotic-supplemented R2A), E. coli (Eosin Methylene Blue), and Pseudomonas spp. (Pseudomonas Isolation Agar). In addition, the tetA gene was quantified from the carrot samples as a measure of the effect of sanitizers and storage on an antibiotic resistance gene known to be carried by the inoculated bacteria.Inclusion of sanitizer in the wash water significantly reduced the absolute numbers of inoculated bacteria (E.coli and Pseudomonas) as well as populations of bacteria capable of growth on the R2A media containing cefotaxime (10μg/mL), sulfamethoxazole (100μg/mL), or tetracycline (3μg/mL). Comparable reductions in the inoculated P. aeruginosa resistant to tetracycline (PIA T, 4μg/mL), bacteria resistant to cefotaxime (10μg/mL) and tetracycline (3μg/mL) occurred after washing with XY-12 or Tsunami 100. The sanitizer effectiveness may be bacterial dependent, as evident by larger absolute reductions of the inoculated E. coli (EMB) and bacteria grown on sulfamethoxazole (100μg/mL)-amended plates after washing with Tsunami 100 compared to washing with tap water or XY-12. Re-growth of both the inoculated and native compost-associated bacteria was inhibited by storage at 2 °C, as there were no significant differences in the log CFU/g values on the various media (total aerobic bacteria, bacteria on antibiotic-amended plates, E. coli inoculum, P. aeruginosa inoculum) during the 14-day storage period. However, temperature abuse at 10 °C resulted in significant re-growth of native Pseudomonas, compared to storage at 2 °C. A sanitizer-associated interaction between re-growth and temperature was also observed for bacteria resistant to clindamycin (25μg/mL) and cefotaxime (10μg/mL), with substantial re-growth occurring only on carrots washed with Tsunami 100. There was no significant re-growth of the inoculated E. coli O157:H7 at either temperature. Results indicate that some bacterial populations are reduced by post-harvest washes and that temperature abuse of fresh produce may result in increases in antibiotic-resistant bacterial populations. / Master of Science in Life Sciences

Prior research has shown pathogen prevalence on-farms is not uniformly distributed, instead pathogen prevalence is highly dependent on environmental factors and management practices. A study was performed to determine environmental factors (e.g., landscape features, meteorological events) and management practices (e.g., date of last irrigation, pesticide application) that may impact the prevalence of Salmonella spp. and Listeria spp. at the sub-field level (0.2 ha grids) on an Eastern Shore of Virginia farm. Virginia Tech's Eastern Shore Agricultural Research and Extension Center (ESAREC) farm was used due to the liability of testing for pathogens in commercial produce fields; however, production practices used at the ESAREC farm are similar, if not the same, to production practices used on commercial farms. Fifteen drag swab, one water, and up to five fecal samples were collected every two weeks per sampling occurrence from August to December 2016 (thus up to 21 samples may be collected during one sampling occurrence. Samples were collected from randomized field plots that were picked during each sampling occurrence. Salmonella spp. and Listeria spp. were isolated and confirmed using modified versions of the Food and Drug Administration's Bacteriological Analytical Manual. Environmental factors were retrieved by remotely-sensed data for the sample location or date. Management practices were recorded by an observational survey for each sample occurrence. Two hundred and seventy-four samples (210 drag swab, 50 fecal, and 14 water samples) were collected during the late summer, fall, and winter. Listeria spp. and Salmonella spp. was detected in 8.3% (23/274) and 1.8% (5/274) of samples, respectively. Neither pathogen was detected in any of the fourteen water samples tested. Findings from this study will support the development of mitigation strategies to reduce pathogen contamination on-farm, with emphasis at the sub-field level. For instance, mitigation strategies include growers electing to not harvest near edges of fields or directly after precipitation events to minimize contamination events. Additionally, management practices were found to be associated with pathogen prevalence; therefore, management practices should be carefully tailored for each unique farm landscape. / Master of Science in Life Sciences

Due to the paucity of long-term harvest impact data, the primary goals of this study were to quantify the long-term effects of different harvest disturbances twenty-four years after harvest on two major wetland functions: stand productivity and C storage. This study evaluated the effects of three harvest types that were originally applied in 1986 to a tupelo (Nyssa aquatic)-cypress (Taxodium distichum) forested wetland in the Mobile-Tensaw River Delta of southwestern Alabama. Treatments were: 1. Helicopter harvest (HELI), 2. Skidder simulation where 50% of the site was rutted to a depth of 30 cm (SKID), and 3. Helicopter harvest followed by glyphosate herbicide removal of all sprouts and seedbank regeneration for two years following harvest (GLYPH). An adjacent mature stand (94 years old) within the same original composition represented mature forest or pre-harvest reference conditions (REF). Above- and belowground plant biomass, belowground woody debris, soil C, and soil CO2 efflux were measured. Twenty-four years after treatments were applied, forest C levels were higher in SKID treatments (206.1 Mg C ha-1) than in HELI treatments (168.7 Mg C ha-1). GLYPH treatments are holding less (144.2 Mg C ha-1) while REF areas hold 332.6 Mg C ha-1. SKID treatments are also holding the most biomass of all treatments with 243.2 Mg ha-1 of overstory biomass. Ecosystem C and biomass patterns indicate HELI and SKID are becoming similar to the original site conditions represented by the REF areas. The resiliency of these highly disturbed stands are explained by the frequent inputs of non-compacted sediments, presence of species well adapted to very poorly drained and aerated conditions, high rates of coppice regeneration, shrink-swell ameliorative properties of the soil and creation of more complex microtopography within SKID treatments. / Master of Science