O emprego de câmeras digitais multiespectrais torna possível a utilização de índices de vegetação, obtidos por meio de operações matemáticas entre bandas espectrais de uma mesma imagem. Estes índices podem ser empregados na estimativa de produtividade de culturas agrícolas e no delineamento de zonas de manejo, por apresentarem relação com o vigor da cultura. Algumas variáveis obtidas no campo, como o índice de área foliar (IAF), a altura de plantas e o número de plantas por metro linear, também podem ser empregadas na avaliação do vigor da cultura. O objetivo principal deste trabalho foi avaliar imagens obtidas por meio de videografia aérea digital multiespectral, quanto ao seu potencial na estimativa da produtividade e na determinação de zonas de manejo em culturas de grãos. As imagens foram adquiridas por meio de uma câmera de vídeo digital multiespectral (Duncantech MS3100). Também foram utilizados mapas de produtividade, referentes a duas áreas cultivadas, primeiramente com trigo, no inverno de 2001, e na seqüência com soja, no verão de 2002. Além disso, foi realizado um trabalho de campo, na cultura da soja, em uma das áreas de estudo, onde foram determinados, em uma grade amostral, o índice de área foliar, a altura de plantas e o número de plantas por metro linear. As imagens aéreas foram corrigidas geometricamente e normalizadas radiometricamente no software Idrisi32, após o que foi realizada uma regressão linear simples entre as imagens e os mapas de produtividade, pixel-a-pixel e com as imagens classificadas. Os dados coletados em grade foram analisados por meio da estatística descritiva e da geoestatística, sendo posteriormente interpolados, gerando os mapas de superfície das variáveis estudadas. Os mapas de superfície criados para as variáveis medidas no campo apresentaram elevada correlação entre sí. A imagem NDVI apresentou uma melhor relação com a estimativa de produtividade, quando comparada com as imagens das bandas espectrais individualizadas, do vermelho e do infravermelho próximo. Concluiu-se que as imagens aéreas digitais multiespectrais obtidas por videografia são eficientes na estimativa da produtividade de grãos quando existe elevada variabilidade nas imagens e as mesmas não apresentam valores discrepantes. Também proporcionam informações importantes ao delineamento de zonas de manejo. / The utilization of multispectral digital cameras makes it possible the use of vegetation indices, generated by means of mathematical operations between spectral bands from the same image. These indices can be used to estimate crop yields and delineate management zones due to the relation between them and crop vigor. Some variables, measured on the field, such as leaf area index (LAI), plants height and plants per linear meter, can also be used in the assessment of the crop vigor. The main objective of this work was to evaluate multispectral digital aerial videographic images regarding their potential in crop yield estimate as well as in management zones delineation. The images were acquired with a multispectral digital camera (Duncantech MS3100). Yield maps were also used, for two cropped areas, firstly with wheat, in the 2001 winter, and afterwards, with soybean, in the 2002 summer. Moreover, a field work was accomplished, for soybean, in one of the study areas. Three variables were measured, on a sampling grid: leaf area index, plants height and plants per linear meter. The aerial images were geometrically rectified and radiometrically normalized, in the software Idrisi32, and then a simple linear regression was performed between images and yield maps, pixel-by-pixel and with the classified images and maps. Data collected on the sampling grid were analyzed by means of descriptive statistics and also geostatistics. After an interpolation procedure, surface maps of these variables were generated. The surface maps generated for the variables measured on the field presented high correlation among themselves. The NDVI image showed a better relation with yield estimate than the individual spectral bands, red and near infrared. One can conclude that the multispectral digital aerial videographic images are efficient for grain crops yield estimates, when there is a high variability on the images, without outliers. These images can also provide important information to the management zones delineation.

Post-fire survival, composition and regeneration of understory species in the boreal forest have shown to be affected by several factors, where consumption of the organic soil layer together with altered soil properties play important parts. There has however also been shown that the pre-fire site characteristics affect the post-fire understory vegetation. This study aimed to investigate the effects of fire and pre-fire site characteristics on understory regeneration and composition at a local scale in a boreal forest. Classification of species richness of the understory species together with measurements of biomass in terms of leaf area index (LAI) and normalized difference vegetation index (NDVI) were performed in a Pinus sylvestris forest in the Gärsjön catchment area, three years after a stand-replacing wildfire. Data of site index, fire severity on soil and moss, fire severity on shrubs, stand age, and remaining humus depth were also used. A total of 36 species of vascular plants (10 forbs, 14 graminoids, 5 dwarf shrubs, 2 ferns, 1 shrub and 4 trees) together with 3 species of bryophytes were recorded in the area. The study revealed that understory species composition was explained by remaining humus depth and site index. The regeneration of the understory was affected differently, where LAI was affected by site index, and NDVI was connected to both site index and fire severity on soil and moss. LAI and NDVI differed in their sensitivity in capturing differences among plant species, where higher values of LAI were associated to species such as E. sylvaticum, P. erecta, C. arundinacea and J. conglomeratus, while NDVI was related to both the ground and field layer, with high values associated to a high abundance of C. canescens and C. ovalis. According to my result, it can be concluded that NDVI is a more appropriate measure of post-fire re-establishment and recovery of understory vegetation in the boreal forest.

Le Sahel, région semi-aride située au sud du désert du Sahara, est particulièrement sensible au changement climatique et est soumise à une forte variabilité inter-annuelle et décennale des précipitations. Au cours des dernières décennies, la région a été marquée par deux périodes de sécheresse intense dans les années 1972-1973 et 1983-1984, qui ont eu des conséquences dramatiques sur les ressources en eau et la végétation, provoquant l'érosion des sols, d'énormes pertes en bétail et des récoltes catastrophiques accentuant la paupérisation des populations rurales. Ces travaux de thèse contribuent à un effort d'identification et de quantification par télédétection des changements d'états de surface sur une grande région du Sahel malien, le Gourma qui s'étend sur près de 90 000 km² et couvre tout le gradient bioclimatique sahélien. L'étude s'appuie sur une importante base de données constituée au sol par des sites pastoraux et agricoles suivis dans le cadre de plusieurs programmes nationaux et internationaux. Trois sources de données de télédétection sont utilisées : 1) l'imagerie multi-spectrale à haute résolution spatiale Landsat, 2) l'imagerie couleur à très haute résolution spatiale observée en vision pseudo-stéréoscopique de Google Earth et 3) le produit NDVI du capteur MODIS à moyenne résolution spatiale mais à haute répétitivité temporelle. Les résultats présentés fournissent une caractérisation détaillée à l'échelle de la région du Gourma des types de surfaces sol-végétation, de l'occupation agricole des sols, du fonctionnement hydrique et aussi des productions végétales moyennes sur 14 années, herbacées et foliaires des ligneux. Pour quelques-unes de ces variables, et sur une partie de la région au moins, les résultats ont quantifié la dynamique historique de 1986 à 2011. Au niveau de la commune de Hombori, l'historique remonte jusqu'à 1955. Les résultats montrent clairement que la dynamique des mises en culture n'est pas responsable de la concentration du ruissellement observé dans le Gourma comme ailleurs au Sahel. / The Sahel is a semiarid region located south of the Sahara desert. The Sahel is particularly sensitive to climatic change and is subject to a high inter-annual and decadal rainfall variability. This region was marked by two periods of very severe droughts which occurred in 1983 - 1984 and 1972-1973 within a long dry period which began in 1970. This long drought deeply impacted water resources and vegetation, causing soil erosion, huge livestock losses and catastrophic harvests accentuating the impoverishment of the rural population. This study is an attempt to characterize and quantify changes in soils, vegetation and hydrology at the scale of the Gourma region (90 000 km²) located in Northern Mali from the last 55 years. Three types of remote sensing data have been used: 1) high spatial resolution Landsat images, 2) very high spatial resolution Geo-Eye images (Google-Earth) and 3) NDVI data measured by the moderate spatial resolution satellite (MODIS). The approach aims to characterize soil surface texture, local redistribution of rainwater by surface run off, surface hydrology and vegetation cover changes over the considered period. Temporal dynamics of vegetation and soils are pointed out. Particularly, this study shows the strong relationships between soil types, surface hydrology and vegetation dynamics at different spatial scales. Land cover changes are also characterized over the considered period.

Understanding the patterns and relationships between land surface productivity and the climatic condition is essential to predict the impact of climate change. This study aims to understand spatial temporal variability and relationships of land surface productivity using Normalized Difference Vegetation Index (NDVI) and drought indices, mainly Standard Precipitation Index (SPI) and Standard Precipitation Evaporation Index (SPEI) across four ecoregions: Mojave, Sonoran, Apache Highlands and Chihuahua of the Southwest United States. Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) and land cover data, and Parameter Regression on Independent Slopes Model (PRISM) precipitation and temperature data were used for analysis. Using Mann-Kendall, I calculated the trends in annual and seasonal NDVI, SPI and SPEI datasets. I used the Pearson Correlation Coefficients to examine the response of integrated and monthly NDVI values to SPI and SPEI values. The positive and negative trends were found during the annual and monsoon seasons whereas only negative trends were found during the spring season for NDVI, SPI and SPEI values. The relationship between NDVI and coincident and antecedent SPEI values changed significantly by area and season for each of the ecoregions across the east-west seasonal precipitation gradient.

Global and local climate patterns may affect barren-ground caribou (Rangifer tarandus groenlandicus) populations. I predicted global climate changes to be correlated with periods of population decline, and local changes to be more pronounced on the habitat of caribou with a declining population. In chapter 1, the Arctic Oscillation (AO), changes in normalized difference vegetation index and phenology were used as measures of global and local climate. In chapter 2 environmental variables and caribou presence points were used to build Maxent habitat models. There was no consistent correlation with the positive AO phase and periods of population decline, or phenology trends and the habitat of caribou with a declining population. Maxent models underestimated the amount of suitable habitat spatially and failed to model suitable habitat temporally. This thesis is the first to look at a range of density-independent variables over a long time period and model suitable habitat for multiple herds.

A large portion of Earth¿s terrestrial surface is subject to arid climatic water stress. As in these regions the hydrological cycle and the vegetation dynamics are tightly interconnected, a coupled modeling of these two systems is needed to fully reproduce the ecosystems¿ behavior over time and to predict possible future responses to climate change. In this thesis, the performance of three parsimonious dynamic vegetation models, suitable for inclusion in an operational ecohydrological model, are tested in a semi-arid Aleppo pine forest area in the south-east of Spain. The first model considered, HORAS (Quevedo & Francés, 2008), simulates growth as a function of plant transpiration (T), evaluating environmental restraints through the transpiration-reference evapotranspiration ratio. The state variable related to vegetation is R, relative foliar biomass, which is equivalent to FAO crop coefficient (Allen et al., 1998), but not fixed in time. The HORAS model was then abandoned because of its unsatisfactory results, probably due to a poor simulation of evaporation and transpiration processes. As for the other two models, WUE-model and LUE-model, the state variable is the leaf biomass (Bl, kg dry mass m-2 vegetation cover). Both models simulate gross primary production (GPP), in the first case as a function of transpiration and water use efficiency (WUE), in the second case as a function of absorbed photosynthetically active radiation (APAR) and light use efficiency (LUE). Net primary production (NPP) is then calculated taking into account respiration. The modelling is focused particularly on simulating foliar biomass, which is obtained from NPP through an allocation equation based on the maximum leaf area index (LAI) sustainable by the system, and considering turnover. An analysis of the information offered by MODIS EVI, NDVI, and LAI products was also performed, in order to investigate vegetation dynamics in the study site and to select the best indices to be used as observational verification for models. MODIS EVI is reported in literature (Huete et al., 2002) to be highly correlated with leaf biomass. In accordance with the phenological cycle timing described for the Aleppo pine in similar climates (Muñoz et al., 2003), the EVI showed maximum values in spring and minimum values in winter. Similar results were found applying the aforementioned WUE- and LUE- models to the study area. Contrasting simulated LAI with the EVI series, the correlation coefficients rWUE = 0.45 and rLUE = 0.57 were found for the WUE-model and LUE-model respectively. Concerning NDVI, its own definition links this index to the ¿greenness¿ of the target, so that it appears highly linked to chlorophyll content and vegetation condition, but only indirectly related to LAI. Photosynthetic pigment concentrations are reported to be sensitive to water stress in Aleppo pine (Baquedano and Castillo, 2006) so, to compare the models¿ results with NDVI, the simulated LAI was corrected by plant water-stress. The resulting correlation coefficients were rWUE = 0.62 and rLUE = 0.59. Lastly, MODIS LAI and ET were found to be unreliable in the study area because very low compared to field data and to values reported in literature (e.g. Molina & del Campo, 2012) for the same species in similar climatic conditions. The performance of both WUE- and LUE- models in this semi-arid region is found to be reasonable. However, the LUE-model presents the advantages of a better performance, the possibility to be used in a wider range of climates and to have been extensively tested in literature. / Pasquato, M. (2013). Comparison of parsimonious dynamic vegetation modelling approaches for semiarid climates [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/34326 / TESIS

No description available.

Potato Virus Y (PVY) infection and nitrogen (N) deficiency cause similar symptoms (chlorosis and stunting) on potato foliage. While conventional methods, including ELISA and petiole testing, require destructive sampling and a longer time to diagnose, spectral analysis can be non-destructive, rapid and efficient. Spectral reflectance for potato cultivars representing three market types, chip processing, red-skinned fresh, and fresh and processing russets, were assessed in separate greenhouse trials in response to three N rates (90, 200, and 290 kg/ha) and two PVYN:O infection levels (clean and infected) at 4, 6, and 8 weeks after inoculation (WAI). Normalized Difference Vegetation Index (NDVI) was able to differentiate clean and PVYN:O infected samples of red-skinned and chip processing cultivars, at 4 and 8 WAI, respectively. Overall, cultivars differed in their spectral responses, indicating the importance of studying cultivar-specific spectral responses against PVY infection in future.

It is widely accepted that climate variability is a key driver of vegetation productivity. Yet, there are discrepancies on the ideal timescales of climatic events and vegetation response. The work herein attempts to clarify how those variables interact in the region of California. The Standard Precipitation Evapotranspiration Index (SPEI), a drought index, was used as an indicator of interannual climate variability in the region. Vegetation productivity was accounted with Normalized Difference Vegetation Index (NDVI) or net growth point data. In this study, four parameters were tested: the length of climate events influencing vegetation, the ideal time to be accounted as vegetation response, the start of the growing season, and the lag between climate and vegetation response.  In total, 594 different scenarios were simulated, with 432 considering the correlation between SPEI and NDVI anomalies and the remaining between SPEI and net growth. The findings shows that the Hot Deserts of California have an early start of the drought season, in March or April, with climate events from 6 months prior influencing vegetation greenness for the next 3 months. In those deserts, the direct correlations between SPEI and NDVI have been the highest, of 0.70 (Mojave) and 0.64 (Sonoran), meaning that, in these locations drier periods will decrease vegetation health. Cold Deserts present a later start of the drought season, in May. Vegetation in these regions will have a delayed response to droughts, with scenarios of 1 to 2 months lags between climate events and vegetation response presenting the highest correlations between SPEI and NDVI. Response that is also longer with climatic events influencing the next 9 months of vegetation greenness. When the correlations were significant, Mediterranean California behaved similarly to cold deserts, with a lag between climate and vegetation, and even longer periods of climatic influence on vegetation, of up to 12 months. In colder regions of California, entailing the entire Western Cordillera, Cold Deserts, and some regions of Mediterranean California an inverse relationship between SPEI and NDVI was found. Drier periods early in the season, in March or April will cause vegetation to be greener during the following months. In cold deserts and Mediterranean California, this climate vegetation relationship happened for short climatic events, as only the previous months will have an impact on vegetation for the following three months. The Cordillera was influenced by longer climatic events, of up to three months, and was the location that showed the best inverse correlations between NDVI and SPEI. In these locations, an early snowmelt and higher temperatures, leading to higher evapotranspiration, could explain the increase in greenness of vegetation by drier periods. However, this observation does not hold when considering a larger scale of climatic events. The correlation between SPEI and Net Growth has showed that when longer periods are considered, with climatic events of 12 or 24 months, a decrease in the net growth of plants will happen for the following season. As Californian climate is predicted to become more extreme it is of great importance understanding the possible consequences for vegetation.

To support the growing human population, plant phenotyping technologies must innovate to rapidly interpret hyperspectral (HS) data into genetic inferences for plant breeders and managers. While pigment and nutrient concentrations within canopies are known to be vertically non-uniform, these chemical distributions as sources of HS noise are not universally addressed in scaling leaf information to canopy data nor in detecting spectral plant health traits. <br>In this project, soybeans (Glycine Max, cultivar Williams 82) were imaged with a Spectra Vista Corporation (SVC) HR-1024 spectroradiometer (350-2500 nm) at the highest five node positions. The samples were subjected to nitrogen and drought stress in factorial design (n=12) that was validated via relative water content (RWC) and PLS Regression of photopigments (chlorophyll a, chlorophyll b, lutein, neoxanthin, violaxanthin, and zeaxanthin in mg/g DW) and N concentration (%) for each imaged tissue. Welch’s ANOVA and Tamhane’s T2 post-hoc testing quantified spectral heterogeneity with respect to treatments and node positions through spectral angle measurements (SAMs) and percent NDVI difference. Drought-stressed samples had the lowest SAM between node positions compared to other treatments, and SAM node comparisons were greatest when including the highest sampled tissues. Taking ratios of NDVI between node positions proved more statistically effective at discerning between all factorial treatments than individual leaf NDVI values. Finally, intra-canopy spectral heterogeneity was exploited by training Linear Discriminant Analysis (LDA) classifiers on relative reflectance between node positions, tuning for the F1-Score. A classifier built on Node 1 vs. Node 3 reflectance outperformed in class-specific accuracies compared to analogous models trained on point-view data. Accounting for intra-canopy spectral variability is an opportunity to develop more comprehensive phenotyping tools for plant breeders in a world with rapidly rising agricultural demand.<br><br>