The influence of habitat loss on biodiversity is related to a set of non-linear processes, which strongly affect isolation and connectivity and determine both extinction and colonization rates. Landscape changes due anthropogenic disturbances are driving not only species loss per se, but also loss of functions performed by those species, which could have important impacts on the provision of ecosystem services and ecosystem function. The objective of this dissertation is to move beyond our current understanding about landscape processes underpinning bird persistence and avian-mediated pest control in fragmented landscapes. In chapter 1, we systematically review the empirical evidence of landscape structure effects on avian-mediated pest control in agricultural systems worldwide in order to point out the main landscape processes underpinning ecosystem service provision and gaps in knowledge where research efforts should be focused. We have found 226 bird species that provide pest regulation worldwide, and a substantial proportion of those are native habitat-dependent species in tropical systems. Moreover, more heterogeneous landscapes, increased habitat amount, and decreased isolation among patches are positively associated with increased avian-mediated pest control rates. Then, using bird data collected in fragmented landscapes in the Brazilian Atlantic forest, in chapter 2 we test for community-level extinction thresholds across a range of biodiversity indices (taxonomic, functional, and phylogenetic diversity) in order to evaluate how much habitat is need to maintain sustainable ecosystem functioning and ecosystem resilience, and how matrix composition might affect these thresholds. In general, we have found that more permeable matrices are able to postpone species loss in fragmented landscapes. However, since habitat loss reaches 20% of habitat remaining, all aspects of biodiversity are compromised. Finally, in chapter 3 we investigate the cross-habitat spillover process - one of the main processes regulating avian-mediated pest control in tropical regions. Matrix composition is a key factor in facilitating species movement into matrices, as 24% of the species pool can spill over into coffee plantations, while spillover into pasture is nearly non-existent. Moreover, the interaction between forest cover and edge density is an important predictor of spillover when habitat amount is low. Our results suggest that landscape configuration and matrix composition should be considered when planning agricultural landscapes in order to ensure long-term persistence of biodiversity and ecosystem services provision / O efeito da perda de habitat sobre a biodiversidade está ligado a um conjunto de processos não-lineares que fortemente afetam a conectividade e o isolamento da paisagem, e determinam taxas de extinção e colonização. Mudanças na estrutura da paisagem desencadeadas por distúrbios antropogênicos levam não somente à perda de espécies per se, mas também a perda de funções que estas espécies desempenham no ecossistema, com importantes implicações em termos de funcionalidade e provisão de serviços ecossistêmicos. O objetivo geral desta tese é elucidar os processos, que ocorrem no nível da paisagem, que modulam a persistência de aves e, consequentemente, as funções destas aves na provisão do controle de pragas em paisagens fragmentadas. No primeiro capítulo da tese, nós revisamos sistematicamente as evidências empíricas dos efeitos da estrutura da paisagem sobre o controle de pragas provido por aves em sistemas agrícolas ao redor do mundo, identificando os processos, no nível da paisagem, subjacentes a este controle e as lacunas de conhecimento onde futuros esforços devem ser concentrados. Identificamos 226 espécies de aves provendo controle de pragas em sistemas agrícolas, sendo que uma porção substancial destas espécies em sistemas tropicais são dependentes de habitats nativos. Em geral, paisagens mais heterogêneas, com elevada cobertura de habitat nativo e menor isolamento entre fragmentos estão positivamente relacionados com elevadas taxas de controle de pragas provido por aves. Por conseguinte, no segundo capítulo, usamos dados de aves em paisagens fragmentadas emersas em diferentes contextos de matriz agrícola na Mata Atlântica brasileira, para testar limiares de extinção ao nível de comunidade. Foram usadas diferentes métricas de diversidade biológica (taxonômica, funcional e filogenética), com o intuito de avaliar o quanto de habitat é necessário para garantir o funcionamento e resiliência do ecossistema. Nós encontramos que matrizes mais permeáveis são capazes de postergar a perda de espécies em paisagens fragmentadas, no entanto, quando a perda de habitat alcança o limiar crítico de 20%, todos os aspectos da biodiversidade estão comprometidos, independentemente do tipo de matriz. Por fim, no capítulo 3, nós investigamos o processo de \'transbordamento\' (i.e. spillover), um dos principais processos reguladores da provisão do serviço de controle de pragas pela biota dependente de habitats nativos. Nós demonstramos que a composição da matriz é um fator determinante facilitando o movimento de espécies para as matrizes agrícolas, sendo que 24% do pool de espécies consegue utilizar matrizes de cafezal, enquanto este movimento é quase inexistente em matrizes de pastagem. Ademais, em paisagens com pouca cobertura florestal, a interação entre cobertura florestal e densidade de borda é um importante preditivo do spillover. Nossos resultados sugerem que a configuração da paisagem e a composição da matriz devem ser consideradas no planejamento de paisagens agrícolas para garantir a persistência em longo prazo da biodiversidade e a provisão de serviços ecossistêmicos

Orientador: Marineide Rosa Vieira / Banca: Marli de Fátima Stradioto Papa / Banca: Mário Eidi Sato / Resumo: A videira é uma cultura de elevada importância econômica, e assim como outras culturas, sofre o ataque de diversas pragas e doenças. Videiras da região do município de Jales, Estado de São Paulo, tem sido intensamente infestadas pelo ácaro-rajado, Tetranychus urticae Koch (Tetranychidae), gerando aumento na utilização de acaricidas. O uso indiscriminado destes acaricidas pode contaminar o meio ambiente, os agricultores e os consumidores dos alimentos. Na busca por um controle alternativo, o presente trabalho teve por objetivo testar o efeito acaricida de produtos naturais sobre os ácaros dessa espécie. Para isso, fêmeas de T. urticae, criadas em laboratório a partir de espécimes coletados em videiras daquele local, foram pulverizadas em torre de Potter. Foram testados três produtos a base de nim (Azadirachta indica Juss.) e extrato pirolenhoso em diferentes concentrações, extratos aquosos e hidroetanólicos de dez plantas e óleos essenciais de seis plantas. Cada tratamento foi testado duas vezes, para confirmação dos resultados. A parcela experimental consistiu de uma placa de Petri com um disco de folha de feijão-de-porco (Canavalia ensiformes L.) contendo cinco fêmeas adultas de T. urticae, utilizando-se dez placas por tratamento. As avaliações foram realizadas após 24, 48, 72, 96 e 120 horas, para contagem do número de fêmeas mortas no disco e fêmeas encontradas no algodão umedecido, utilizado como barreira. As maiores mortalidades foram registradas nos tratamentos com Natuneem, óleo emulsionado de sementes de nim, extrato hidroetanólico de camomila (Matriacaria chamomilla L.) e extrato aquoso de erva doce (Pimpinella anisum L.). Óleos essenciais não tiveram efeito acaricida sobre T. urticae. Extratos aquosos de hortelã (Mentha x piperita L.), melissa (Melissa officinalis L.) e losna (Artemisia absinthium L.) e hidroetanólicos... (Resumo completo, clicar acesso eletrônico) / Abstract: The vine is a culture of high economic importance, and as other cultures, it is attacked by various pests and diseases. Vines of the municipality of Jales, Sao Paulo State, there has been heavy infestations by two spotted spider mite, Tetranychus urticae Koch (Tetranychidae), generating an use of acaricides increased. The indiscriminate use of acaricides can contaminate the environment, farmers and food consumers. In the search for an alternative control, this study aimed to test the acaricidal effect of natural products on mites of this species. For this, T. urticae females, reared in the laboratory from specimens collected in vines that place, were sprayed in tower Potter. We tested three products based on neem (Azadirachta indica Juss.) and pyroligneous acid in different concentrations, aqueous and hydroethanolic extracts of ten plants and essential oils of six plants. Each treatment was tested twice to confirm the results.The experimental plot consisted of a Petri dish with a Canavalia ensiformis L. leaf disk containing five T. urticae female, using ten plates per treatment. The evaluations were realized after 24, 48, 72, 96 and 120 hours, with counting of females killed in the disk and in cotton wool barrier. The highest mortalities were recorded in Natuneem, emulsified oil from neem seeds, chamomile (Matricaria chamomilla L.) aqueous extract and fennel (Pimpinella anisum L.) hydroethanolic extract treatments. Essential oils have no acaricide effect on T. urticae. High mortalities were recorded with peppermint (Mentha x piperita L.), lemon balm (Melissa officinalis L.) and wormwood (Artemisia absinthium L.) aqueous extracts and with pacari (Lafoensia pacari Saint-Hilaire) and oregano (Origanum vulgare L.) hydroethanolic extract... (Complete abstract click electronic access below) / Mestre

nÃo hà / O dengue permanece como importante problema de saÃde pÃblica e a principal estratÃgia de controle empregada permanece sendo o uso de larvicidas contra as formas imaturas de seu principal vetor, o mosquito Aedes aegypti. Por conta das limitaÃÃes existentes com o uso desses larvicidas à importante a busca por outras alternativas de controle. O uso de peixes larvÃfagos, como mecanismo de controle biolÃgico, tem sido incentivado e desta forma, o objetivo deste trabalho foi avaliar o comportamento de oviposiÃÃo de fÃmeas gravidas de Aedes aegypti na presenÃa de peixes larvÃfagos e a possibilidade de sua utilizaÃÃo em consÃrcio com os larvicidas recomendados pelo Programa Nacional de Controle do Dengue (PNCD). Foi adaptada a fÃrmula de Kramer e Mulla (1979) para cÃlculo do Ãndice de Atividade de OviposiÃÃo (IAO), com trÃs diferentes situaÃÃes. Foram testadas trÃs espÃcies de peixe (Betta splendens, Trichogaster tricopterus e Poecilia reticulata) e trÃs diferentes larvicidas (Bti, novaluron e temefÃs). Para determinar a capacidade de repelÃncia utilizamos uma gaiola com volume de 6m3, em que foram colocados nove tambores com capacidade para 70 litros de Ãgua. Em trÃs tambores foi colocado somente Ãgua (controle), em outros trÃs um peixe e nos outros trÃs o larvicidas a ser testado. ApÃs uma semana os mosquitos postos nessa gaiola eram retirados e os ovos contados com auxÃlio de um microscÃpio entomolÃgico. Cada associaÃÃo peixe/larvicida foi repetida 5 vezes, perfazendo um total de 30 experimentos. Para avaliar a sobrevivÃncia dos peixes utilizou-se 15 caixas dâÃgua com 250 litros de Ãgua. Em 10 caixas foi colocado o peixe com o respectivo larvicida a ser testado e em outras cinco, somente o peixe (controle). Cada associaÃÃo peixe/larvicida foi repetida 4 vezes. Os peixes T. tricopterus e P. reticulata nÃo apresentaram nenhuma influÃncia na atividade de oviposiÃÃo das fÃmeas de A. aegypti, o mesmo acontecendo com os larvicida novaluron e temefÃs. Jà o peixe B. splendens apresentou repelÃncia à postura de ovos, enquanto o larvicida Bti atraiu a postura por parte das fÃmeas grÃvidas. Todos os peixes avaliados sobreviveram aos larvicidas recomendados pelo PNCD, sugerindo ser viÃvel sua utilizaÃÃo em consÃrcio, em programas integrados de controle de vetores. / Dengue remains an important public health problem and the main control strategy employed remains the use of larvicides against the immature forms its principal vector , the mosquito Aedes aegypti. Because of limitations with the use of these larvicides is important to search for other alternatives control. The use of larvivorous fish, such as biological control mechanism, has been encouraged in the world and thus the objective of this study was to evaluate the oviposition behavior of pregnant females of Aedes aegypti in the presence of larvivorous fish and usability in consortium with larvicides recommended by the Program National Control Dengue (PNCD). Was adapted formula Kramer and Mulla (1979) for calculation of Oviposition Activity Index (OAI), with three different situations. We tested three species of fish (Betta splendens , Trichogaster tricopterus and Poecilia reticulata) and three different larvicides (Bti , novaluron and temefos). To determine the repellency use a cage with a volume of 6m3, where drums were placed nine with 70 liters of water. Three barrels were placed in water only (control), one in three other fish and other three larvicides to be tested. After a week the mosquitoes put this cage eggs were removed and counted with the aid of a microscope entomological. Each association fish / larvicide was repeated 5 times for a total of 30 experiments. To evaluate the survival of fish we used 15 water tanks with 250 liters of water. In 10 cases the fish was placed with its larvicide to be tested and in five, only the fish (control). Each association fish / larvicide was repeated 4 times. Fish T. tricopterus and P. reticulata showed no influence on oviposition activity of females of A. aegypti, the same happening with the larvicide temefos and novaluron. Already the fish B. splendens showed repellency to oviposition while the larvicide Bti attracted posture on the part of pregnant females. All fish survived the evaluated recommended larvicides PNCD, suggesting feasible to use in consortium programs in integrated vector control.

In Western Australia dried meat baits containing 1080 are used extensively by agricultural and conservation organisations to control foxes and dingoes for the protection of agricultural production and native fauna. Field trials were conducted to assess 1080 loss from dried meat baits and this required the analysis of over five hundred baits. Because of this large number of baits it was essential to have a simple and efficient 1080 extraction procedure and method of 1080 analysis. In this study three methods of 1080 extraction and the new bioassay method for 1080 analysis were investigated. A simple and cost-effective 1080 extraction method using water with a 98% 1080 recovery rate was developed and modifications to the bioassay method were made.Factory-produced 1080 dried meat baits were laid in the field during different seasons at four locations in Western Australia, samples were collected over time and analysed for 1080 content using the bioassay. Rainfall was recorded and temperature data was collected for each site. Baits were exposed to the elements but were placed in mesh or wire cages to restrict invertebrate attack and prevent removal by vertebrates. Some baits were placed on the surface and others were buried. Initially 1080 loss from baits from all 4 sites was minimal, ranging from 0 - 21% at day 7 - 9. Further loss was gradual even when rainfall was recorded. Generally baits had to be exposed to at least 50 mm of rain before 1080 loss increased to 50%. At some sites baits continued to remain toxic to foxes even after long exposure. The mean 1080 content of baits from the Carnarvon site at day 226 was 2.0 mg (55% of the mean 1080 content of baits at day zero) with 137 mm of rainfall recorded for that period. Loss of 1080 from baits buried occurred at a faster rate than from baits placed on the surface during the same time period. By day 14 no 1080 was ++ / detected in the buried baits compared to the 68% detected in the surface baits. Under certain conditions 1080 loss from baits was minimal. Levels of 1080 in baits from Nangeen Hill remained fairly constant during the months of September to December 1995, and again during February to April 1996.Gastrolobium plant tissue and soil samples from the southwest of Western Australia were investigated for the presence of 1080 degrading microorganisms. Microbes were isolated and individually tested in solution containing 1080 as the sole carbon source. Isolates which showed 1080 degrading ability were further tested for their degrading efficiency in McClung carbon-free solution with added 1080 as the sole source of carbon and in factory 1080 waste solution, at 1080 concentrations of 20 and 200 mM. The effect of temperature on their rate of degradation was also examined. Thirteen isolates (7 fungi and 6 bacteria) showing varying degrees of 1080 degrading ability were obtained. Rates of 1080 degradation varied among isolates but were highest in the factory waste solution at the 20 mM concentration and in the McClung solution, where 1080 was the sole source of carbon, at the higher concentration of 200 mM. The most efficient isolates OSK and 10H (both Pseudomonas species) degraded all the 1080 present in sterile factory waste solution up to 20 mM 1080 concentration in 4 days and the isolate 1AF (Fusarium oxysporum) degraded 93% of 200 mM 1080 in the McClung solution in 9 days. The optimum temperatures for 1080 degradation were 30 degrees celsius and fluctuating ambient temperatures of 15 28 degrees celsius.

Many candidate mycoherbicides have shown promise in the laboratory or greenhouse, but most have been ineffective in the field. Factors limiting mycoherbicide efficiency include temperature and humidity. Results from this thesis indicate that solar radiation has both a damaging effect(reduction in germination)limiting efficacy and a photomorphogenic effect(appressorium induction)increasing efficacy. The study has also shown significant interaction between temperature and solar radiation on the survival of conidia of potential mycoherbistats. Therefore, solar radiation should be considered as third major component of the environment that should be considered when trying to produce mycoherbistats. With the findings presented in this thesis and further research on disease development under different conditions, in combination with the formulation of conidia in suitable UV protectants, a computer simulation modelling the conditions leading to epidemics caused by C.orbiculare, D.avenacea and R.alismatis could be constructed. It may be possible to manipulate fungal application time in order to expose conidia to doses of solar radiation that are not harmful to conidium germination and which stimulate appressorium formation. However, additional protection may be needed. / Doctor of Philosophy (PhD)

We identified and evaluated a phytoseiid predator as a biological control agent of multiple spider mites pests that occur in ornamental nurseries. When comparing species, Neoseiulus fallacis (Garman) had a wider prey range than Galendromus occidentalis (Nesbitt), a higher probability of overwintering than Neoseiulus califomicus McGregor and was equally or more effective at suppressing spider mites than either of the other 2 phytoseiids in 4 field tests. To further evaluate N. fallacis we 1) measured prey range when held with 29 ornamental pests or alternative foods under laboratory conditions, 2) tested biological control of spider mites on representative plant species at both small and large spatial scales, 3) developed release and conservation strategies of the predator, and 4) examined the efficacy of the predator in controlling recently introduced pests. Neoseiulus fallacis had greatest survival and reproduction when feeding on spider mites but eriophyid mites, other mites and pollen enhanced survivorship and, in some cases, reproduction. When inoculated into ornamental plants, spider mite suppression was "acceptable" in 81% of small scale tests and-in all large scale tests. Limitations in control occurred in tall, vertical growing plants with little foliar canopy. Inoculation of N. fallacis at low prey densities into apple rootstocks was successful at suppressing Tetranychus urticae Koch and similar to control achieved at moderate prey densities. In small scale banker plant studies, high densities of adult and immature mites of N. fallacis were produced and moved downwind to receiver plants. In field tests with receiver plants placed at greater distances, only N. fallacis adult females readily dispersed to 30 m or more. When comparing overwintering survival of adult females among plant types, N. fallacis survived most on conifers, intermediate on evergreen shrubs and least on herbaceous perennials, deciduous shrubs and shade trees. Covering plants with protective plastic reduced overwintering survival of the predator. Neoseiulus fallacis successfully suppressed the newly introduced pests Panonychus citri (McGregor) and Schizotetranychus celarius (Banks) on Skimmia japonica Thunberg and Sasaella hidaensis (Makino and Uchida), respectively. Initial studies suggest that N. fallacis can be an effective biological control agent of multiple spider mites in low-growing and selected higher-growing ornamental plants. / Graduation date: 2000

The life history, temperature, humidity, and feeding requirements of the phytoseiid predator mite, Typhlodromus americanus Chant and Yoshida-Shaul were tested in the laboratory. Occurance of the mite in the field during different times of the year was investigated as well. Effects of temperature and humidity on egg hatch, the feeding requirements of the larvae, the amount consumed by each life stage, the length of each life stage and suitability of different food sources were investigated in the laboratory. The use of the mite as a biological control agent was evaluated by using the information gained from the laboratory experiments. T. americanus was originally discovered in plantation grown Douglas-fir in western Oregon. Since that time the mite has been found on a number of other hosts throughout North America. The mite is active year round in the Christmas tree plantations of the Willamette Valley in Oregon. The adult is found in or near the one year old bud scars and the eggs are typically deposited there also. Life parameters were measured providing a net reproductive rate of 4.23, a mean generation time of 24.45 days, and an intrinsic rate of increase of 0.059 The intrinsic rate of increase was low when compared to other predator mites and numerical response to prey increase would not be possible with such a low rate. The optimal temperature for the shortest eclosion time (54.4 hr.) and the highest survival (96.4%) was 26��C. The regression of temperature vs. time to hatch gave a 90% R�� with both the slope and intercept significantly different from zero. Humidities above 70% had survival rates over 96% and eclosion rates in the range of 50-58 hours. The relative humidity at which 50% of the population died was 58.6%. The mite was found to feed readily on the pest mite Oligonychus ununguis (spruce spider mite), as well as Tetranychus urticae (two spotted spider mite), and corn, oak, and Douglas-fir pollens. The larval form of the predator mite does not require food to molt to the protonymph, but the protonymph does require food to molt. If water is provided the entire time from egg to death, the protonymph can survive about ten days. Females consumed more Tet. urticae than males in both the immature and adult stages. The T. americanus that were fed corn pollen and Tet. urticae (complete diet) lived for over 115 days. Mites raised on oak and corn pollens did not survive as long (only 70-80 days), and those raised on Douglas-fir pollens did not reach adulthood. Egg production was observed on the complete diet, but not on the diets of pollen. The largest number of eggs were laid around the twelfth day after the molt to adult. Control and management of field conditions to improve habitat for T. americanus will be the best approach for its use as a biological control agent. As it does not respond numerically to prey increase, it will be more effective in a regulatory role to prevent these increases while the prey is at low levels. / Graduation date: 1996

Hippodamia convergens Guerin-Meneville was selected for rapid development through five generations at four constant temperatures (18, 22, 26, and 30��C). Two levels of selection were used: intense and moderate. Differences in developmental rate, survivorship, live adult weight, aphid consumption, adult longevity, and fecundity were measured for both groups and an unselected line. The two selected lines exhibited significant differences in developmental rate after the first generation of selection. The differences increased by the fifth generation of selection. Average differences between selected lines at 18, 22, 26, and 30��C in development from egg to adult were 4.9, 3.0, 1.0, and 1.5 days, respectively. Individuals from the intense selected line had a lower developmental threshold (11.3��C) than the moderate selected line (12.0��C). Also, degree-day requirements from egg to adult were lower in the intense (266 DD��) than moderate (277 DD��) line. Wild type beetles accumulated 231 degree-days above a threshold of 13.6��C. Survivorship at cold temperatures significantly increased with selection for fast development. In addition, no significant differences occurred in either live adult weight, total larval consumption of prey, fecundity, or adult longevity. Larvae selected for fast growth consumed higher numbers of aphids per day and were more efficient metabolically in converting prey mass into body mass than slow growing larvae. Because of a lower developmental threshold and lower degree-day requirements, for complete development, beetles from the intense selected line may accumulate more degree-days throughout a year compared to a moderate selected line. Predictions of population growth, based on an equation for intrinsic growth rate, showed that beetles from the intense selected line would produce 1.6 and 1.4 times more individuals in a 60-day period than beetles from the moderate selected and unselected lines, respectively. The results suggest that intense selected beetles would develop faster earlier in the season and would produce higher population numbers under optimum conditions. Thus, selection of H. convergens produced superior beetles for future introductions in biological control. / Graduation date: 1995

Thesis (M.S.)--University of Kentucky, 2003. / Title from document title page (viewed June 21, 2004). Document formatted into pages; contains xii, 122 p. : Ill. Includes abstract and vita. Includes bibliographical references (p. 106-121).

4 pp. / Commercially available sonic pest devices for use in residential applications have not been shown to be effective in scientific studies. For this reason, use of these devices is not advised to treat common pest problems. Although some researchers are developing sonic techniques that illustrate promise for very specific pests, these technologies are yet to be commercially available. As our understanding increases of how pest species receive and process sound, more relevant sonic devices may be developed. The allure of sound as a treatment for pests will remain into the future—motivated by the fact that if they are successful they will be more environmentally friendly and safer for humans.