Tese (doutorado)—Universidade de Brasília, Instituto de Ciências Biológicas, Departamento de Botânica, Programa de Pós-graduação em Botânica, 2014. / Submitted by Patrícia Nunes da Silva (patricia@bce.unb.br) on 2015-12-18T13:50:35Z
No. of bitstreams: 1
2014_JulianaSilvestreSilva_Parcial.pdf: 37400751 bytes, checksum: 3dd00dd131a19bc77177c6871280dc30 (MD5) / Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2016-01-05T13:38:37Z (GMT) No. of bitstreams: 1
2014_JulianaSilvestreSilva_Parcial.pdf: 37400751 bytes, checksum: 3dd00dd131a19bc77177c6871280dc30 (MD5) / Made available in DSpace on 2016-01-05T13:38:37Z (GMT). No. of bitstreams: 1
2014_JulianaSilvestreSilva_Parcial.pdf: 37400751 bytes, checksum: 3dd00dd131a19bc77177c6871280dc30 (MD5) / Em ambientes tropicais as espécies vegetais apresentam padrões fenológicos influenciados por fatores bióticos e abióticos que podem ser estudados a partir de exsicatas depositadas em herbário. A informatização dessas coleções vem tornando os materiais herborizados cada vez mais utilizados, fundamentando estudos com os mais variados objetivos, principalmente em formações vegetais sob alta pressão antrópica, como o Cerrado. Os herbários são grandes bancos de dados que guardam informações geográficas, taxonômicas, ecológicas e fenológicas de espécies variadas e esses registros têm sido uma ferramenta útil para examinar tendências em longo prazo dos eventos fenológicos, principalmente quando relacionadas às mudanças temporais. Nos últimos anos, estudos fenológicos vêm sendo adotados com frequência como forma de prever o comportamento de sistemas biológicos frente às mudanças no clima, visto que entender o que controla a fenologia da vegetação é essencial na tentativa de compreender as possíveis mudanças futuras em relação à mudança de tempo e clima. Sabe-se que mudanças não previstas nos padrões climáticos estão ocorrendo e o aumento da temperatura tem interferido na distribuição geográfica, fisiologia, fenologia e adaptação in situ das espécies, assim como alterado o ciclo hidrológico, causando mudanças nos padrões de precipitação. Um dos objetivos deste trabalho foi desenvolver um índice matemático capaz de prever o comportamento fenológico de espécies vegetais a partir de registros contidos em grandes bancos de dados e como resultado foi incorporado ao programa BRAHMS um algoritmo interno denominado Phenological Predictability Index (PPI), que calcula a probabilidade de ocorrência dos eventos fenológicos em campo, com base em exsicatas, literatura e imagens datadas. Paralelamente, foi realizado o monitoramento quinzenal em campo, no Distrito Federal, de 22 espécies no período entre 2011 e 2013, das quais 11 nunca foram estudadas no Distrito Federal. Foram incluídas nas análises outras sete espécies, previamente estudadas na mesma região, entre os anos de 2000 e 2003. Os estudos de campo permitiram o cálculo da constância do evento fenológico, necessário para se deduzir a contingência, e o poder preditivo do PPI como ferramenta de planejamento de expedições de campo foi testado. Observaram-se quatro padrões fenológicos vegetativos e quatro padrões de antese e produção de botões e frutos. Para os eventos de floração o PPI previu os meses de maior atividade de cerca de 90% das espécies analisadas e para frutificação a previsão do mês-pico ocorreu para pouco menos de 50% delas. O período em que a fenofase tem maior chance de ocorrer foi previsto corretamente para 100% das espécies nos três eventos analisados (botão, flor e fruto). Considerando a existência da variação geográfica na fenologia sugere-se que os registros herborizados usados para o cálculo do PPI estejam dentro da mesma região climática dos locais onde ocorrerá o estudo ou a coleta. O insucesso no uso do índice poderá ocorrer no caso de espécies com fenologia supra-anual ou com baixo valor de PPI, o que indica alta contingência do evento, isto é, plasticidade fenotípica da espécie frente às mudanças locais ou climáticas entre anos. Sabendo da existência dessas mudanças interanuais no clima, diferentes técnicas têm sido utilizadas para estudar as variabilidades climáticas e seus efeitos, além de definir e prever padrões e anomalias. Assim, este trabalho propôs também o uso da morfometria geométrica, mais especificamente a técnica Thin-Plate Spline (TPS), no estudo de variáveis climáticas – em escala local e num período de 50 anos – para que dados climáticos de diversas décadas pudessem ser comparados entre si e os anos climaticamente semelhantes fossem identificados. O uso da morfometria geométrica em estudos voltados à variabilidade climática é inédito e a técnica se mostrou bastante útil ao apresentar a variabilidade anual existente entre os decêndios do ano, mesmo essa variação sendo muito pequena. Foi possível identificar que as variáveis temperatura máxima e precipitação apresentaram maior variação interanual no período chuvoso, enquanto para a variável temperatura mínima a maior variação ocorreu no período seco. Os agrupamentos obtidos apresentaram-se fracos para as três variáveis, no entanto, a análise reconheceu anos com presença de El Niño e La Niña no estudo da precipitação, variável que apresentou a maior variabilidade entre anos. A teoria da probabilidade foi utilizada para explorar o grau de previsibilidade e de oportunismo da floração no Distrito Federal e investigar possíveis relações com características climáticas como temperatura, precipitação e fotoperíodo. Como conclusão, obteve-se o resultado comportamental mais amplo e geral de diferentes e importantes plantas do Cerrado e provou-se que o uso de espécimes coletados em um grande período de tempo – como registros de herbários – é útil no estudo da fenologia vegetal, quando se tem como objetivo definir comportamentos e padrões frente às mudanças ambientais. A identificação da variabilidade climática em pequeno espaço de tempo (p.ex. decêndios do ano), como feito por meio da técnica de Thin-Plate Spline, é de grande importância em estudos de vegetação, assim como a distinção de anos com ocorrência dos fenômenos La Niña e El Niño. Verificou-se que para a obtenção de melhores resultados, a análise da morfometria deve ser restrita a períodos mais curtos de tempo, como a estação chuvosa, quando ocorrem maiores variações climáticas no Distrito Federal, ou nos meses que antecedem a ocorrência dos eventos fenológicos. Poucas espécies parecem independer de algum estímulo ambiental para florescer, a maioria delas parece responder a uma combinação complexa de estímulos como precipitação, temperatura e fotoperíodo, combinados a ritmos endógenos e, para algumas, à passagem do fogo. A utilização do PPI como forma de quantificar a chance que o material de interesse tem de ser encontrado em campo tende a garantir o sucesso das expedições de campo, resultando em aceleração e economia para as pesquisas. / In tropical vegetation, flowering plants have phenological patterns influenced by biotic and environmental factors and that can be studied using herbarium specimens. Herbarium databasing has accelerated the use of such data, providing baseline information for studies with many kinds of different objectives, particularly for regions under pressure from humans, such as the Cerrado. Herbaria are large databases that preserve geographic, taxonomic, ecologic and phenologic information and such records are very useful to study long-term tendencies in phenology. In the last few years, phenological studies have become very popular as a key to predicting how biological systems will respond to climate changes, since understanding what controls plant phenology is essential to try and understand possible future changes in climate and weather. It is becoming clear that unprecedented changes in climatic patterns are underway and that the increase in temperature is interfering in the distribution, physiology, phenology and in situ adaptation of many species, as well as modifying the hydrological cycle, therefore changing rainfall patterns. One of the objectives of this study was to develop a mathematical index capable of predicting the phenological behaviour of plant species based on large databases. As a result, a formula was devised and an algorithm was incorporated into BRAHMS that calculates an index we have called the Phenological Predictability Index (PPI), that calculates the probability of registering a phenological event in the field based on herbarium, literature and dated images. Our study focused on a representative sample of Angiosperm species that are common the Distrito Federal, Brazil. Along with phenological scoring of herbarium data, 22 species were monitored twice a month in the field between 2011 and 2013. Seven species that were studied between 2000 e 2003 also in the Distrito Federal were included in the analyses. Our field study (as well as other published studies) permitted us to calculate average individual constancy of the phenological event so that contingency could be deduced. Results obtained in the field were compared with those predicted by the PPI index. Peak flowering month was congruent with maximum monthly probability for c. 90% of species and peak fruiting month was congruent with maximum monthly probability in 50% of species. Scoring events in the field was 100% successful for visits in peak probability months (buds, flowers and fruit). Considering the existence of geographic variability in phenology, we suggest that the herbarium records used to calculate PPI are within the same climatic region where the study or collecting expedition is planned to occur. Lack of success in PPI prediction of an event may be due to the fact that it is supra-anual (mast species) or event predictability is very low, meaning it is highly plastic, presumably responding to year-to-year climatic variation or to local conditions. A second aim of this study is to improve knowledge of patterns and identify possible anomalies in year-to-year climatic variability. This study used geometric morphometrics, specifically the Thin-Plate Spline (TPS) technique, in the study of decendial climatic variables – minimum temperature, maximum temperature and precipitation – on a local scale and over a time scale of 50 years, thus allowing climatic data for several decades to be compared and putatively climatically similar years to be identified. The use of geometric morphometrics in climate studies is unprecedented and the technique was useful to quantify the decendial climatic variability even when it was small. Precipitation showed the highest variability by far with the highest value found in the second decendium of November. Maximum temperatures showed maximum variability during the wet season and minimum temperatures during the dry season. UPGMA was used to group years using the three variables, but bootstrap values were below 50, suggesting 50 years might be insufficient for pattern formation. However, UPGMA based on precipitation was successful in grouping El Niño and La Niña years. Probability theory was used to explore the degree of flowering predictability and opportunism in the Distrito Federal and investigate possible relationships with climatic variables, such as temperature, precipitation and photoperiod. We conclude that the behaviour of this sample of a diversified and important group of common Cerrado plants studied – as herbarium records – is useful in the study of plant phenology, if the objective is to elucidate behaviour and patterns in response to environmental variables. The identification of climatic variability using short time (such as 10 days), as done using the Thin-Plate Spline technique, is of great importance in vegetation studies, as is the distinction of years with La Niña and El Niño phenomena. For best results, morphometric analyses should be restricted to seasons, such as the rainy season, when major climatic variability occurs in the Distrito Federal, or in the months that preceed the desired phenological events. Few species appear to be independent of environmental stimulus to flower, and many of them may respond to a complex pattern of stimuli such as precipitation, temperature and photoperiod, combined with endogenous rhythms or after burning. Using PPI as a means of quantifying the probability that botanical material has of being found the desired phenological state in the field can enhance success in field expeditions, resulting in faster and more economical research. / In tropical vegetation, flowering plants have phenological patterns influenced by biotic and environmental factors and that can be studied using herbarium specimens. Herbarium databasing has accelerated the use of such data, providing baseline information for studies with many kinds of different objectives, particularly for regions under pressure from humans, such as the Cerrado. Herbaria are large databases that preserve geographic, taxonomic, ecologic and phenologic information and such records are very useful to study long-term tendencies in phenology. In the last few years, phenological studies have become very popular as a key to predicting how biological systems will respond to climate changes, since understanding what controls plant phenology is essential to try and understand possible future changes in climate and weather. It is becoming clear that unprecedented changes in climatic patterns are underway and that the increase in temperature is interfering in the distribution, physiology, phenology and in situ adaptation of many species, as well as modifying the hydrological cycle, therefore changing rainfall patterns. One of the objectives of this study was to develop a mathematical index capable of predicting the phenological behaviour of plant species based on large databases. As a result, a formula was devised and an algorithm was incorporated into BRAHMS that calculates an index we have called the Phenological Predictability Index (PPI), that calculates the probability of registering a phenological event in the field based on herbarium, literature and dated images. Our study focused on a representative sample of Angiosperm species that are common the Distrito Federal, Brazil. Along with phenological scoring of herbarium data, 22 species were monitored twice a month in the field between 2011 and 2013. Seven species that were studied between 2000 e 2003 also in the Distrito Federal were included in the analyses. Our field study (as well as other published studies) permitted us to calculate average individual constancy of the phenological event so that contingency could be deduced. Results obtained in the field were compared with those predicted by the PPI index. Peak flowering month was congruent with maximum monthly probability for c. 90% of species and peak fruiting month was congruent with maximum monthly probability in 50% of species. Scoring events in the field was 100% successful for visits in peak probability months (buds, flowers and fruit). Considering the existence of geographic variability in phenology, we suggest that the herbarium records used to calculate PPI are within the same climatic region where the study or collecting expedition is planned to occur. Lack of success in PPI prediction of an event may be due to the fact that it is supra-anual (mast species) or event predictability is very low, meaning it is highly plastic, presumably responding to year-to-year climatic variation or to local conditions. A second aim of this study is to improve knowledge of patterns and identify possible anomalies in year-to-year climatic variability. This study used geometric morphometrics, specifically the Thin-Plate Spline (TPS) technique, in the study of decendial climatic variables – minimum temperature, maximum temperature and precipitation – on a local scale and over a time scale of 50 years, thus allowing climatic data for several decades to be compared and putatively climatically similar years to be identified. The use of geometric morphometrics in climate studies is unprecedented and the technique was useful to quantify the decendial climatic variability even when it was small. Precipitation showed the highest variability by far with the highest value found in the second decendium of November. Maximum temperatures showed maximum variability during the wet season and minimum temperatures during the dry season. UPGMA was used to group years using the three variables, but bootstrap values were below 50, suggesting 50 years might be insufficient for pattern formation. However, UPGMA based on precipitation was successful in grouping El Niño and La Niña years. Probability theory was used to explore the degree of flowering predictability and opportunism in the Distrito Federal and investigate possible relationships with climatic variables, such as temperature, precipitation and photoperiod. We conclude that the behaviour of this sample of a diversified and important group of common Cerrado plants studied – as herbarium records – is useful in the study of plant phenology, if the objective is to elucidate behaviour and patterns in response to environmental variables. The identification of climatic variability using short time (such as 10 days), as done using the Thin-Plate Spline technique, is of great importance in vegetation studies, as is the distinction of years with La Niña and El Niño phenomena. For best results, morphometric analyses should be restricted to seasons, such as the rainy season, when major climatic variability occurs in the Distrito Federal, or in the months that preceed the desired phenological events. Few species appear to be independent of environmental stimulus to flower, and many of them may respond to a complex pattern of stimuli such as precipitation, temperature and photoperiod, combined with endogenous rhythms or after burning. Using PPI as a means of quantifying the probability that botanical material has of being found the desired phenological state in the field can enhance success in field expeditions, resulting in faster and more economical research.
Identifer | oai:union.ndltd.org:IBICT/oai:repositorio.unb.br:10482/19079 |
Date | 03 December 2014 |
Creators | Silva, Juliana Silvestre |
Contributors | Proença, Carolyn Elinore Barnes |
Source Sets | IBICT Brazilian ETDs |
Language | Portuguese |
Detected Language | Portuguese |
Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/doctoralThesis |
Source | reponame:Repositório Institucional da UnB, instname:Universidade de Brasília, instacron:UNB |
Rights | A concessão da licença deste item refere-se ao termo de autorização impresso assinado pelo autor com as seguintes condições: Na qualidade de titular dos direitos de autor da publicação, autorizo a Universidade de Brasília e o IBICT a disponibilizar por meio dos sites www.bce.unb.br, www.ibict.br, http://hercules.vtls.com/cgi-bin/ndltd/chameleon?lng=pt&skin=ndltd sem ressarcimento dos direitos autorais, de acordo com a Lei nº 9610/98, o texto integral da obra disponibilizada, conforme permissões assinaladas, para fins de leitura, impressão e/ou download, a título de divulgação da produção científica brasileira, a partir desta data., info:eu-repo/semantics/openAccess |
Page generated in 0.0028 seconds