Desenvolveu-se nesse trabalho uma metodologia para identificar e quantificar quinzenalmente a área queimada na região contínua do Cerrado brasileiro a partir de imagens diárias do sensor AVHRR (Advanced Very High Resolution Radiometer) do satélite NOAA-14 (National Oceanic and Atmospheric Administration), utilizando as bandas 1 (0.6 um), 2 (0,9 um), 3 (3.7 um) e IVDN (Índice de Vegetação de Diferença Normalizada). A variação temporal de características das áreas queimadas e outras superfícies, obtida de mosaicos quinzenais das imagens AVHRR, foi a base para elaborar um algoritmo de identificação de \"cicatrizes\" de queimadas. Os resultados foram validados e ajustados com dados de alta resolução espacial obtidos em imagens TM/Landsat (Thematic Mapper). A análise estatística de regressão linear entre os dados de queimadas obtidos pela aplicação do algoritmo nos mosaicos AVHRR e aqueles das imagens TM gerou duas equações para estimar a área queimada no Cerrado com r2 = 0,8 e 0,7. Com a aplicação da metodologia desenvolvida, estimou-se em ~429.000 km2 a área queimada (entre 404.000 km2 e 455.000 km2 com intervalo de confiança a 95%) no período de 01/maio/98 a 30/abril/99, correspondendo a 19% (18 a 20%) da área total estudada. A relação entre focos de queimadas obtidos do AVHRR/NOAA-12 e área queimada permitiu cálculos preliminares de área queimada no Cerrado no período de 01/maio/99 a 31/outubro/00. Os dados TM mostraram que as queimadas pequenas, menores que 0,5 km2, embora muito numerosas (53% do total), respondem por apenas ~2 % da área queimada. Por outro lado, as queimadas grandes, maiores que 10 km2, são poucas (8%), mas responsáveis por cerca de 74% da área queimada no Cerrado. Os resultados desse trabalho mostraram pela primeira vez que é possível estimar regularmente a área queimada no Cerrado com erro inferior a 15% no cálculo anual a partir dos dados diários do AVHRR. Tais estimativas poderão subsidiar estudos sobre o papel ecológico do fogo no Cerrado, planejamento ambiental em nível regional, localização das áreas críticas com ocorrências mais freqüentes de queimadas, implantação de planos de uso, manejo e fiscalização do uso do fogo em escala regional, cálculos de emissões de queimadas, etc. / This work presents the development of a methodology to identify and quantify the surface burnt in the Brazilian contiguous Cerrado on a bi-weekly basis using daily images of the AVHRR (Advanced Very High resolution Radiometer) sensor on-board the NOAA-14 (National Oceanographic and Atmospheric Administration) satellite after its bands 1 (0,6 um), 2 (0,9 um) and 3 (3,7 um), as well as the NDVI (Normalized Difference Vegetation Index). The temporal variation of the burnt areas and of other surface covers in bi-weekly AVHRR mosaics was the basis for an algorithm to identify the \"scars\" from vegetation fires. The results were validated and adjusted with high resolution data from TM-Landsat (Thematic Mapper). The statistical analysis of linear regression between the fire data obtained with the use of the algorithm and those of the TM produced two equations to estimate burnt area in the Cerrado, with r2 = 0.8 and 0.7. Applying the methodology developed, ~429,000 km2 burned in the period of May/01/98 to April/04/99 (range of 404,000 to 455,000 km2 for the 95% confidence interval), corresponding to 19% (18 to 20 %) of the total study area. The relation between active fires obtained with AVHRR/NOAA-12 and the burnt area supplied preliminary estimates of burnt area in the Cerrado from May/01/99 to Oct/31/00. The TM data showed that small scars, with less than 0.5 km2, although numerous (53 % of the total), account for just ~2 % of the burnt area. Large scars, with more than 10 km2, correspond to a small number (8 %), but to 74 % of the Cerrado burnt area. The results of this work showed for the first time that it is possible to estimate on a regular basis the Cerrado yearly burnt area with an error smaller than 15 %, using daily AVHRR data. These estimates should provide important information to understand the ecological role of fire in the Cerrado, identify areas with higher fire frequency, help environmental planning at regional levels, and plan soil use and control, as well as provide subsidies in biomass burning emission studies. Data from new sensors in satellites to be made available in 2001 should improve even further the methodology developed.

Desenvolveu-se nesse trabalho uma metodologia para identificar e quantificar quinzenalmente a área queimada na região contínua do Cerrado brasileiro a partir de imagens diárias do sensor AVHRR (Advanced Very High Resolution Radiometer) do satélite NOAA-14 (National Oceanic and Atmospheric Administration), utilizando as bandas 1 (0.6 um), 2 (0,9 um), 3 (3.7 um) e IVDN (Índice de Vegetação de Diferença Normalizada). A variação temporal de características das áreas queimadas e outras superfícies, obtida de mosaicos quinzenais das imagens AVHRR, foi a base para elaborar um algoritmo de identificação de \"cicatrizes\" de queimadas. Os resultados foram validados e ajustados com dados de alta resolução espacial obtidos em imagens TM/Landsat (Thematic Mapper). A análise estatística de regressão linear entre os dados de queimadas obtidos pela aplicação do algoritmo nos mosaicos AVHRR e aqueles das imagens TM gerou duas equações para estimar a área queimada no Cerrado com r2 = 0,8 e 0,7. Com a aplicação da metodologia desenvolvida, estimou-se em ~429.000 km2 a área queimada (entre 404.000 km2 e 455.000 km2 com intervalo de confiança a 95%) no período de 01/maio/98 a 30/abril/99, correspondendo a 19% (18 a 20%) da área total estudada. A relação entre focos de queimadas obtidos do AVHRR/NOAA-12 e área queimada permitiu cálculos preliminares de área queimada no Cerrado no período de 01/maio/99 a 31/outubro/00. Os dados TM mostraram que as queimadas pequenas, menores que 0,5 km2, embora muito numerosas (53% do total), respondem por apenas ~2 % da área queimada. Por outro lado, as queimadas grandes, maiores que 10 km2, são poucas (8%), mas responsáveis por cerca de 74% da área queimada no Cerrado. Os resultados desse trabalho mostraram pela primeira vez que é possível estimar regularmente a área queimada no Cerrado com erro inferior a 15% no cálculo anual a partir dos dados diários do AVHRR. Tais estimativas poderão subsidiar estudos sobre o papel ecológico do fogo no Cerrado, planejamento ambiental em nível regional, localização das áreas críticas com ocorrências mais freqüentes de queimadas, implantação de planos de uso, manejo e fiscalização do uso do fogo em escala regional, cálculos de emissões de queimadas, etc. / This work presents the development of a methodology to identify and quantify the surface burnt in the Brazilian contiguous Cerrado on a bi-weekly basis using daily images of the AVHRR (Advanced Very High resolution Radiometer) sensor on-board the NOAA-14 (National Oceanographic and Atmospheric Administration) satellite after its bands 1 (0,6 um), 2 (0,9 um) and 3 (3,7 um), as well as the NDVI (Normalized Difference Vegetation Index). The temporal variation of the burnt areas and of other surface covers in bi-weekly AVHRR mosaics was the basis for an algorithm to identify the \"scars\" from vegetation fires. The results were validated and adjusted with high resolution data from TM-Landsat (Thematic Mapper). The statistical analysis of linear regression between the fire data obtained with the use of the algorithm and those of the TM produced two equations to estimate burnt area in the Cerrado, with r2 = 0.8 and 0.7. Applying the methodology developed, ~429,000 km2 burned in the period of May/01/98 to April/04/99 (range of 404,000 to 455,000 km2 for the 95% confidence interval), corresponding to 19% (18 to 20 %) of the total study area. The relation between active fires obtained with AVHRR/NOAA-12 and the burnt area supplied preliminary estimates of burnt area in the Cerrado from May/01/99 to Oct/31/00. The TM data showed that small scars, with less than 0.5 km2, although numerous (53 % of the total), account for just ~2 % of the burnt area. Large scars, with more than 10 km2, correspond to a small number (8 %), but to 74 % of the Cerrado burnt area. The results of this work showed for the first time that it is possible to estimate on a regular basis the Cerrado yearly burnt area with an error smaller than 15 %, using daily AVHRR data. These estimates should provide important information to understand the ecological role of fire in the Cerrado, identify areas with higher fire frequency, help environmental planning at regional levels, and plan soil use and control, as well as provide subsidies in biomass burning emission studies. Data from new sensors in satellites to be made available in 2001 should improve even further the methodology developed.

Made available in DSpace on 2016-06-02T19:29:43Z (GMT). No. of bitstreams: 1 TeseACP.pdf: 5193455 bytes, checksum: 5bc13e84b2705d44928f3d7b974a269a (MD5) Previous issue date: 2002-11-14 / Universidade Federal de Sao Carlos / At the present, the Cerrado (a type of the Brazilian savannas) is burned on about 20 to 30% of its area during the dry season mainly owing to anthropic causes. Three simultaneous factors are needed for the burnings to happen: favourable meteorological conditions; availability of vegetation fuel; existence of an ignition source. This work studied the susceptibility of the Cerrado vegetation to fire with respect to parameters linked to the three factors: rainfall, relative air humidity and air temperature with respect to the meteorological conditions; vegetation coverage classes with respect to the biomass fuel; proximity to the roads and fire spots from previous days with respect to the ignition source. Location data for the fire spots obtained from the AVHRR/NOAA-12 channel 3 (3,7 mm) images were used as field truth. The study period was between May and October 1998. The study area was divided into 50 km x 50 km cells. The meteorological conditions occurring in 95% of the cells presents fire spots were: rainfall lower than 2 mm; 5-day cumulative precipitation lower than 25 mm; relative air humidity lower than 60%; air temperature higher than 28oC; more than one rainless day before the burning. More than 80% of the Cerrado were susceptible to the fire occurrence, with both locations with and without fire spots presented the minimum meteorological conditions favourable to the vegetation burning described in the literature: rainfall lower than 5 mm; 5-day cumulative precipitation lower than 20 mm; relative air humidity lower than 60%; air temperature higher than 25oC; A method for classifying the Cerrado vegetation coverage to fire susceptibility was also developed. This method was based on 2-weekly mosaics of the AVHRR/NOAA-14 Normalized Difference Vegetation Index (NDVI) and of the channel 3 images. Seven classes of vegetation coverage were discriminated and associated to four degrees of susceptibility: very low, low, medium and high. It was verified that 72% of the burning occurred in the high and medium susceptibility classes, which indicated satisfactory results on the preliminary development of this method. Lastly, the distance between the fire spots and two indicators of anthropic activity was analysed. The indicators were: roads and fire spots previously occurred. About a quarter of the fire spots occurred at up to 10 km from the roads in a area of 582,000 km2 surrounding the roads, which is about 27% of the Cerrado´s total area. On the same way, a quarter of the spots occurred at up to 10 km from the fire spots of the previous day, in an average area of 33,000 km2 at the burning spots surroundings. This area is about 2% of the total area of the Cerrado. In conclusion, the indicators of anthropic activity analysed here area good tools for studying the vegetation susceptibility to fire. / Anualmente, o Cerrado é queimado em 20 a 30% de sua área durante a estação seca, principalmente devido a causas antrópicas. Três fatores simultâneos são necessários para que as queimadas ocorram: condições meteorológicas propícias; disponibilidade de combustível vegetal; existência de fonte de ignição. Este trabalho estudou a susceptibilidade da vegetação do Cerrado ao fogo em relação a parâmetros desses três fatores: precipitação, umidade relativa e temperatura do ar, em relação às condições meteorológicas; classes de cobertura vegetal, em relação ao combustível vegetal; proximidade de malha viária e de focos de queimadas dos dias anteriores, em relação à fonte de ignição. Como verdade de campo foram utilizados os dados de localização dos focos de queimadas obtidos de imagens do canal 3 (3,7 mm) do Advanced Very High Resolution Radiometer / National Oceanic and Atmospheric Administration - 12 (AVHRR/NOAA-12). O período de estudo foi de maio a outubro/1998. A área de estudo foi dividida em células de 50 km por 50 km. As condições meteorológicas em 95% das células com queimadas foram: precipitação inferior a 2 mm; precipitação acumulada de 5 dias inferior a 25 mm; umidade relativa do ar inferior a 60%; temperatura do ar superior a 28oC; mais de um dia sem chuva antecedendo a queimada. Mais de 80% do Cerrado estavam susceptível a ocorrência de fogo, com os locais com e sem focos de queimadas apresentando as condições meteorológicas mínimas propícias à combustão da vegetação relatadas na literatura: precipitação inferior a 5 mm; precipitação acumulada de 5 dias inferior a 20 mm; umidade relativa do ar inferior a 60%; temperatura do ar superior a 25oC. Um método para classificar a cobertura vegetal do Cerrado quanto à susceptibilidade ao fogo também foi desenvolvido, baseado em mosaicos quinzenais do Índice de Vegetação por Diferença Normalizada (IVDN) e do canal 3 gerados com base em imagens AVHRR/NOAA-14. Sete classes de cobertura vegetal foram discriminadas, as quais foram associadas a quatro graus de susceptibilidade: muito baixo, baixo, médio e alto. Foi verificado que 72% dos focos de queimadas ocorreram nas classes de susceptibilidade alta e média, indicando resultados satisfatórios no desenvolvimento preliminar desse método. Por último, foi analisada a distância entre focos de queimadas e dois indicadores de atividade antrópica: malha viária e focos de queimadas recentemente ocorridos. Cerca de um quarto dos focos de queimadas ocorreram até 10 km da malha viária, em uma área ao longo das vias de 582 mil km2, cerca de 27% da área total do Cerrado. Do mesmo modo, um quarto dos focos ocorreu até 10 km dos focos de queimadas ocorridos no dia anterior, em uma área média no entorno dos focos de 33 mil km2, cerca de 2% do Cerrado. Portanto, a proximidade de indicadores de atividade antrópicas pode ser um bom instrumento para avaliação da susceptibilidade da vegetação ao fogo.

Orientador: Hilton Silveira Pinto / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Agricola / Made available in DSpace on 2018-08-06T03:42:16Z (GMT). No. of bitstreams: 1 Almeida_CarlosAlbertoSoaresde_D.pdf: 685356 bytes, checksum: 3eb22116a96bd0f7b5643e5620f19c26 (MD5) Previous issue date: 2001 / Resumo: Foram utilizadas séries de imagens diurnas e noturnas captadas pelos sensores AVHRR a bordo do satélite NOAA-14 para a obtenção dos valores finais de temperatura de superfície do dossel vegetativo. O processamento das imagens incluiu o geo-referenciamento e as correções quanto aos efeitos atmosféricos e de emissividade, para as quais se utilizou o método de correção split window. A equação de split window usada considerou o parâmetro de emissividade da superfície cultivada estimada diretamente no campo pelo método da caixa. Este método foi comparado a um método alternativo de estimativa de emissividade no campo baseado na proporção de cobertura vegetal e solo descoberto. Os dois métodos apresentaram resultados equivalentes quanto à estimativa de emissividade da superfície. A alta correlação entre a temperatura de superfície medida no campo e a obtida a partir das imagens indicou que as equações de split window usadas corrigiram satisfatoriamente as imagens quanto aos efeitos atmosféricos e de emissividade / Abstract: Diurnal and nocturnal image series have been used in this work. These images were processed on suitable software to the treatment of NOAA-AVHRR products, for Surface temperature final value obtaining. The processing work includes both the images geo-referencing and their atmospheric correction applying the "split window method". The split window equation used takes into account the crop surface emissivity parameter, estimated directly on the field by the "box method". This one was compared to an alternative emissivity estimate on the field method, which is based on the crop covering/bared soil relation. Both methods when compared presented similar results. A terrestrial radiometer was used to the surface temperature measurements carried out in situ, during the satellite passes. The correct surface temperature satellite data were submitted to a regression analysis against the terrestrial data about the same parameter. High association level between surface temperature measured at the field and the near air temperature was observed. The high correlation between surface temperature measured at the field and that one from the images shown that the split window equation corrected satisfactorily the atmosphere effects over the images / Doutorado / Agua e Solo / Doutor em Engenharia Agrícola

La productivité primaire nette PPN des grands écosystèmes est la variable-clé pour comprendre et quantifier les intéractions biosphère / atmosphère. Le modèle d'efficience photosynthétique de Monteith mettant en relation la PPN avec la quantité de rayonnement solaire photosynthétiquement actif absorbé (ou intercepté) APAR par le couvert végétal, mesurable à partir des indices de végétation, se prête parfaitement à une approche par télédétection. Le coefficient qui lie ces deux paramètres est appelé rendement photosynthétique ou encore efficience de conversion aérienne nette notée En. L'incertitude sur la valeur de En et l'absence de valeur pour nombre de biomes montrent l'importance de l'étalonnage du modèle. On dispose de mesures de biomasse permettant de calculer la PPN pour 60 des régions écologiques du Québec (qui en compte 69), d'une longue série temporelle d'images satellitaires de Global Vegetation Index NOAA-AVHRR de 1985 à 1992 et des normales climatiques au Québec. La mise en évidence de corrélation entre les propriétés optiques de la végétation (liées à l'APAR) et sa croissance montre la faisabilité d'un suivi de cette PPN par satellite. La biomasse aérienne anhydre est de 7 kg/m2 pour l'ensemble du territoire du Québec. La PPN varie de 100 à 400 g de matière sèche aérienne/m2/an pour le domaine de la pessière du Nord, à 670 g/m2/an pour les domaines feuillus du Sud. Les paramètres qui régissent la productivité forestière ne sont pas constants sur tout le Québec. Le modèle original est donc validé par région écologique appartenant à un même sous-domaine et pour des conditions climatiques données. L'étude de la sensibilité de En à ces variables climatiques a permis d'élaborer un tableau de correspondance entre les valeurs d'APAR, variant entre 664 et 1342 MJ/m2, des variables climatiques d'une région et sa productivité. Ainsi, En varie de 0,1 g/Mégajoules pour le domaine de la pessière noire à 0,57 g/MJ pour les feuillus du sud de la forêt boréale. Pour affiner ces estimations, l'accent devra être mis sur la connaissance des phénomènes physiques qui régissent les propriétés optiques de la végétation et surtout sur une harmonisation des méthodes de calcul et de mesure.

El objetivo general del presente trabajo de investigación fue analizar la variabilidad de la temperatura satelital superficial del mar (TSM) y su efecto en las principales pesquerías que operan sobre un sector de la Plataforma Continental Argentina, que abarca desde el Golfo San Jorge (GSJ) hasta el Talud Continental (TC) adyacente. Esta zona es estratégica dentro del Proyecto Nacional Pampa Azul por su importancia socio-ambiental. Para abordar dicho objetivo se elaboró, por primera vez, una matriz de puntos equidistante que permitió ensamblar productos satelitales derivados de los sensores NOAA/AVHRR y Aqua-MODIS. Esta matriz de puntos posibilitó compararlos (período de solapamiento 2003-2006) y analizar la factibilidad de utilizar una combinación de ambos productos para construir series de tiempo más extensas y confiables. El estudio comparativo de la TSM demostró un alto grado de correlación entre ambos sensores, evidenciando que los productos satelitales son similares entre sí y que pueden ser utilizados de manera conjunta. A partir de los resultados obtenidos se construyó una serie satelital de 35 años de información radiométrica de la TSM proveniente de los sensores NOAA/AVHRR y Aqua-MODIS, la cual permitió realizar un análisis de la variabilidad de este parámetro y sus correspondientes anomalías. Además, se analizaron sus fluctuaciones con el fin de identificar tendencias de cambio térmico en el área de estudio. Para ello, y dadas las características batimétricas y oceanográficas diversas que actúan sobre el área de estudio, se dividió al mismo en tres sectores: Zona 1: Golfo San Jorge; Zona 2: Plataforma Continental (PC) y Zona 3: Talud continental. La señal de la TSM evidenció en todo el sector una variabilidad significativa intra-estacional, anual e interanual (28 meses). Sin embargo, se identificaron señales con diferentes rangos de periodicidad dependiendo de la Zona (1, 2 o 3) estudiada. Posteriormente, se examinó la variabilidad de la anomalía de la TSM (ATSM) y sus posibles conexiones con los siguientes índices climáticos: Modo Anular del Sur (SAM), Índice de la Oscilación del Sur (SOI) y Índice Niño Oceánico (ONI). El análisis permitió establecer que las variaciones atmosféricas remotas influyen en la ATSM local. Si bien se ratifica el efecto del SAM sobre el área, se destacó el efecto del ONI-SOI durante las últimas décadas. El vínculo entre las variaciones de la TSM y las capturas del Abadejo (Genypterus blacodes), Calamar argentino (Illex argentinus), Langostino (Pleoticus muelleri) y Merluza común (Merluccius hubbsii) en combinación con las presiones de origen antrópico (valor de la especie en el mercado y el valor del combustible) permitió determinar que la variación de las capturas de ciertas especies como el Abadejo y el Calamar argentino están relacionadas de manera significativa con la variación de la TSM. El análisis demostró que los registros de las capturas por unidad de esfuerzo (CPUE) del Abadejo se asociaron a meses en que la TSM estuvo por encima de valor promedio. Para el Calamar argentino, el aumento de la CPUE fue posterior a los registros de ATSM positivos. En cambio, para el caso del Langostino y la Merluza, los resultados permitieron determinar que los valores de CPUE no mostraron relación con la variación de la temperatura superficial del mar. Asimismo, posibilitó entender que los forzantes antrópicos ejercen presión sobre el aumento de las capturas ya que en las cuatro pesquerías se observó una correlación significativa entre el valor de la especie y el valor del combustible. Para la pesquería del Abadejo, el Calamar y la Merluza esta relación fue positiva, indicado que los precios de estas especies han ido aumentando a lo largo del período analizado, reflejando un equilibrio entre el costo y el beneficio. En cambio, para el Langostino está correlación fue negativa, evidenciando un comportamiento inverso en la relación costo-beneficio de dicha pesquería. Los resultados obtenidos permitieron determinar el grado de complejidad que conlleva entender las relaciones atmósfera-océano-sociedad, sino que también sentó las bases para comprender que los factores que influyen en una determinada pesquería no necesariamente se aplican a otras. / ABSTRACT The purpose of the present research work was to analyze the variability of satellite sea surface temperature (SST) and its effect on the main fisheries operating on a sector of the Argentine Continental Shelf (ACS), which extends from the San Jorge Gulf (SJG) to the adjacent Continental Slope (CS). This area is strategic within the Pampa Azul National Project due to its socio-environmental importance. For the first time, an equidistant grid was developed, which allowed the assembly of satellite products derived from NOAA/AVHRR and Aqua-MODIS sensors. This grid made it possible to compare them (overlap period 2003-2006) and analyze the feasibility of using both products to build longer and more reliable time series. The SST comparative study showed a high degree of correlation between both sensors, evidencing that the satellite products are similar and can be used jointly. Based on the obtained results, a satellite series of 35 years of SST radiometric information from NOAA/AVHRR and Aqua-MODIS sensors was constructed, which allowed an analysis of the variability of this parameter and its corresponding anomalies. In addition, its fluctuations were analyzed to identify thermal change trends in the study area. For this purpose, and given the different bathymetric and oceanographic characteristics acting on the study area, it was divided into three sectors: Zone 1: SJG; Zone 2: ACS and Zone 3: CS. The SST signal showed significant intra-seasonal, annual and interannual (28 months) variability throughout the area. However, signals with different periodicity ranges were identified depending on the Zone (1, 2 or 3) studied. Subsequently, the variability of the SST anomaly (SSTA) and its possible connections with the Southern Annular Mode (SAM), the Southern Oscillation Index (SOI) and the Oceanic Niño Index (ONI) were examined. The analysis allowed establishing that remote atmospheric variations influence the local SSTA. Even though the effect of SAM on the area was confirmed, the effect of ONI-SOI during the last decades was highlighted. The link between SST variations and catches of pollock (Genypterus blacodes), Argentine squid (Illex argentinus), shrimp (Pleoticus muelleri) and common hake (Merluccius hubbsii) in combination with pressures of anthropic origin (the market value of the species and the value of fuel) allowed determining that the variation in catches of certain species such as pollock and Argentine squid are significantly related to SST variation. The analysis showed that catch per unit effort (CPUE) records for pollock were associated with months when SST was above the average value. For Argentine squid, the increase in CPUE followed positive SST records. On the other hand, in the case of shrimp and hake, the results allowed us to determine that CPUE values showed no relationship with SST variation. Likewise, it was possible to understand that anthropic forcing exerts pressure on the increase in catches since in the four fisheries, a significant correlation was observed between the value of the species and the value of the fuel. For the pollock, squid and hake fisheries, this relationship was positive, which indicates that the prices of these species have been increasing throughout the period analyzed, reflecting a balance between cost and benefit. On the other hand, this correlation was negative for shrimp, showing an inverse behavior in the cost-benefit relationship of this fishery. The obtained results made it possible to determine the degree of complexity involved in understanding the atmosphere- ocean-society relationships and laid the foundations for understanding that the factors that influence a particular fishery do not necessarily apply to others. / TEXTO PARCIAL en período de teletrabajo

In order to understand the dynamics of desertification after the 1980s droughts, the trends and changes of photosynthetically active fraction of the vegetation cover of the semi-arid region of West Africa have been analyzed across three ecoclimatic entities. This study focuses on six countries (Senegal, Mauritania, Mali, Burkina-Faso, Niger and Benin) that reflect quite well the human and environmental context of semi-arid regions. The main objective of this thesis is to check in the before-mentioned biophysical and human context, if the state of the vegetation improves because of more favourable rainfall conditions, or if, conversely, the long environmental deterioration during recent decades has a healthy dose of irreversibility for several years. The process of investigation proceeds in three stages defined by geographical scales and a geoclimatic gradient. At the regional level, we compared vegetation productivity data from 1982-1999 time series of NOAA-AVHRR NDVI data to rainfall data. We analyzed data from 315 rain gauges distributed across five countries (Senegal, Mauritania, Mali, Burkina-Faso and Niger) with annual average isohyets ranging from 100 to 900 mm. For trends analysis, we used the ratio of the integrated vegetation index (iNDVI) during the crop-growth period (June to October) to the May to October sum of rainfall (RR). This ratio (iNDVI/RR), a proxy of the Rain Use Efficiency, is widely used by ecologists as an indicator for detecting desertification processes. Overall, our results show a significant increase of the net primary production as a response of post-drought rainfall increase. However, the trends of iNDVI/RR ratio suggest that most of the studied stations (54%) in sahelian and sahelo-sudanese belts were stable and 31.4% showed strong to very strong negative change in iNDVI/RR ratio, while 13.3% of the stations showed a moderate decrease and only 1.3% showed a positive trend. At the country level, similar trends were recorded throughout 128 stations in Burkina Faso located between the 500 mm and 1100 mm isohyets. In fact, more than half the studied stations showed a stability of iNDVI/RR ratio. However, 39.8% of them show a negative trend from low to high, while only 2.4% of them show a slight positive trend. In addition, a comparison with more detailed local case studies confirmed these observed trends. However, the obtained results for wetter stations in the southern part of the country should be taken with precaution, as the relationship between NDVI and rainfall tends to weaken when annual rainfall is higher than 1000 mm. Overall, the negative trends highlight a gradual decline in plant productivity. These results recorded in 44.7% of the analyzed stations may reflect ongoing desertification processes in the sahelian and sahelo-sudanian zones over the past two decades. At the local level, a structural characterization of woody units in three bioclimatic regions of the sudanian zone (900 mm 1200 mm) in the north of Benin was conducted to assess their degradation status. We recorded the structural characteristics of stands (basal area, density), species diversity as well as disturbances type and intensity. Multivariate analysis revealed a gradient of productivity between the three regions: there was a high diversity of woody stands in the south-sudanian sector, while the north-sudanian and sudano-sahelian sectors were dominated by savannas and shrub, which had low productivity. The productivity gradient is influenced by a disturbance gradient suggesting that the decline in productivity is stronger from south-sudanian to the sudano-sahelian region. In addition, the spatial component of the observable changes in vegetation cover has been mapped by remote sensing in a restricted area of the sudanian zone in northern Benin with SPOT-XS data over the period 1986-2005. Over the past two decades, 19.6% of the woody stands have completely disappeared; 12.9% of this extinction of woody stands was due to deforestation, and 13.9 % due to degradation processes. In contrast to these trends, 21.8% of the study area were stable, while less than a third (31.7%) of the area were experiencing woody recovery (reforestation). The analysis also reveals significant disparities in the rates of change of the identified land use class units. These variations are more pronounced for the woody units and agricultural land than in villages. These results suggest that land cover degradation throughout the study area is primarily due to anthropogenic factors (livestock and agricultural expansion, logging, breeding). In fact, this area is a preferred destination for agricultural migrants fleeing the unpredictable climatic conditions of the drier semi-arid areas. Overall, our results highlight the rapid decline of vegetation resources, challenging assumptions that the impact of ongoing desertification processes is mixed, outside of the arid and semi-arid regions of West Africa. The developed framework is easily reproducible and allows the implementation of a reliable diagnosis on the state of the vegetation cover from accessible and inexpensive data. Its implementation should facilitate the development of managerial techniques for better assistance to the poor, especially vulnerable to the gradual degradation of their environment.