Rapid population growth in Arizona is leading to increasing demand and decreasing availability of water, requiring a detailed quantification of hydrological processes. The integration of detailed spatial water fluxes information from remote sensing platforms, and hydrological models is one of the steps towards this goal. One example step is the use of MODIS Snow Cover Area (SCA) information to update the snow component of a land surface model (LSM). Because cloud cover obscures the images, this project explores a rule-based method to remove the clouds. The rules include: combination of SCA maps from two satellites; time interpolation method; spatial interpolation method; and the probability of snow occurrence in a pixel based on topographic variables. The application in sequence of these rules over the Upper Salt River Basin for WY 2005 resulted in a reduction of cloud obscuration by 93.7878% and the resulting images' accuracy is similar to the accuracy of the original SCA maps. The results of this research will be used on a LSM to improve the management of reservoirs on the Salt River. This research seeks to improve SCA data for further use in a LSM to increase the knowledge base used to manage water resources. It will be relevant for regions were snow is the primary source of water supply.

Reanalysis products are widely used to study the land-atmosphere exchanges of energy, water, and carbon fluxes, and have been evaluated using in situ data above or below ground. Here measurements for several years at five flux tower sites in the U.S. (with a total of 315,576 hours of data) are used for the coupled evaluation of both below- and above-ground processes from three global reanalysis products and six global land data assimilation products. All products show systematic errors in precipitation, snow depth, and the timing of the melting and onset of snow. Despite the biases in soil moisture, all products show significant correlations with observed daily soil moisture for the periods with unfrozen soil. While errors in 2 meter air temperature are highly correlated with errors in skin temperature for all sites, the correlations between skin and soil temperature errors are weaker, particularly over the sites with seasonal snow. While net shortwave and longwave radiation flux errors have opposite signs across all products, the net radiation and ground heat flux errors are usually smaller in magnitude than turbulent flux errors. On the other hand, the all-product averages usually agree well with the observations on the evaporative fraction, defined as the ratio of latent heat over the sum of latent and sensible heat fluxes. This study identifies the strengths and weaknesses of these widely-used products, and helps understand the connection of their errors in above- versus below-ground quantities.

Accurate areal measurements of snow-cover extent are important for hydrological and climate modeling. The traditional method of mapping snow cover is binary where a pixel is approximated to either snow-covered or snow-free. Fractional snow cover (FSC) mapping achieves a more precise estimate of areal snow-cover extent by determining the fraction of a pixel that is snow-covered. The two most common FSC methods using Moderate Resolution Imaging Spectroradiometer (MODIS) images are linear spectral unmixing and the empirical Normalized Difference Snow Index (NDSI) method. Machine learning is an alternative to these approaches for estimating FSC, as Artificial Neural Networks (ANNs) have been used for estimating the subpixel abundances of other surfaces. The advantages of ANNs over the other approaches are that they can easily incorporate auxiliary information such as land-cover type and are capable of learning nonlinear relationships between surface reflectance and snow fraction. ANNs are especially applicable to mapping snow-cover extent in forested areas where spatial mixing of surface components is nonlinear. This study developed an ANN approach to snow-fraction mapping. A feed-forward ANN was trained with backpropagation to estimate FSC from MODIS surface reflectance, NDSI, Normalized Difference Vegetation Index (NDVI) and land cover as inputs. The ANN was trained and validated with high spatial-resolution FSC derived from Landsat Enhanced Thematic Mapper Plus (ETM+) binary snow-cover maps. ANN achieved best result in terms of extent of snow-covered area over evergreen forests, where the extent of snow cover was slightly overestimated. Scatter plot graphs of the ANN and reference FSC showed that the neural network tended to underestimate snow fraction in high FSC and overestimate it in low FSC. The developed ANN compared favorably to the standard MODIS FSC product with the two methods estimating the same amount of total snow-covered area in the test scenes.

There is an undisputed need to increase accuracy of snow cover estimation in regions comprised of complex terrain, especially in areas dependent on winter snow accumulation for a substantial portion of their annual water supply, such as the Western United States, Central Asia, and the Andes. Presently, the most pertinent monitoring and research needs related to alpine snow cover area (SCA) are: (1) to improve SCA monitoring by providing detailed fractional snow cover (FSC) products which perform well in temporal/spatial heterogeneous forested and/or alpine terrains; and (2) to provide accurate measurements of FSC at the watershed scale for use in snow water equivalent (SWE) estimation for regional water management. To address the above, the presented research approach is based on Landsat Fractional Snow Cover (Landsat-FSC), as a measure of the temporal/spatial distribution of alpine SCA. A fusion methodology between remotely sensed multispectral input data from Landsat TM/ETM+, terrain information, and IKONOS are utilized at their highest respective spatial resolutions. Artificial Neural Networks (ANNs) are used to capture the multi-scale information content of the input data compositions by means of the ANN training process, followed by the ANN extracting FSC from all available information in the Landsat and terrain input data compositions. The ANN Landsat-FSC algorithm is validated (RMSE ~ 0.09; mean error ~ 0.001-0.01 FSC) in watersheds characterized by diverse environmental factors such as: terrain, slope, exposition, vegetation cover, and wide-ranging snow cover conditions. ANN input data selections are evaluated to determine the nominal data information requirements for FSC estimation. Snow/non-snow multispectral and terrain input data are found to have an important and multi-faced impact on FSC estimation. Constraining the ANN to linear modeling, as opposed to allowing unconstrained function shapes, results in a weak FSC estimation performance and therefore provides evidence of non-linear bio-geophysical and remote sensing interactions and phenomena in complex mountain terrains. The research results are presented for rugged areas located in the San Juan Mountains of Colorado, and the hilly regions of Black Hills of Wyoming, USA.

The objective of this research is to develop a system capable of simulating snow depth and snow water equivalent in the Sherbrooke to Mount-Megantic area of Quebec's Eastern Townships using meteorological and digital terrain data as input.The working hypothesis is that meteorological data may drive a point energy and mass balance snow cover model.The model used was developed by the Hydrologic Research Lab (National Weather Service) which was calibrated for local conditions using field data collected during two winters at several sites on Mount-Megantic. Snow water equivalent and depth are used for calibration and validation of the model. Automated snow sensors were also used to obtain temperature calibration data.The snow surveys and correction of the air temperature for elevation improves the estimates of snow depth and water equivalent.The results suggest that data from the Sherbrooke meteorological stations can be used to estimate the snow cover over the area of Eastern Townships. Air temperature extrapolation across the field area is a challenge. However the simulated snow cover conforms generally well with data observed at several stations throughout the region.

Résumé : Cette recherche présente l'analyse spatio-temporelle de l'évolution conjointe de l'augmentation de la présence arbustive et de la dynamique de la fonte de la neige au Nunavik, Québec, Canada. Cette zone est caractérisée par la complexité de l'interaction de multiples changements simultanés de la température, de la couverture de la neige ainsi que de la pousse végétative. La première partie de ce travail consiste à faire l'analyse de l’évolution temporelle de ces multiples changements. Cette analyse a nécessité la mise en place d’une importante base de données climatiques, satellitaires et de couverture de sol à plusieurs échelles, sur une période allant jusqu'à 60 ans, soit de 1950-2012. La deuxième partie du travail consiste à faire l'analyse spatiale à haute résolution de l’influence de la fraction du couvert arbustif sur la fonte de la neige. L'analyse et l'interprétation des résultats obtenus dans la première partie montrent clairement un changement climatique significatif sur la région étudiée, découpée en 3 bandes de latitude correspondant à la toundra, la taïga ouverte et à la taïga forestière, respectivement du Nord au Sud. Ce changement de climat correspond à un réchauffement marqué, entre 0.75°C et 1.57°C par décade entre les zones 1 (toundra) et 3 (taïga forestière) respectivement. On peut noter que la hauteur de neige maximale annuelle a diminué dans les trois zones alors que les précipitations hivernales ont augmenté en zone 1 et 3 sur les 45 dernières années. Les résultats montrent une nette augmentation de la végétation arbustive dans les zones 2 et 3 (LAI plus élevé de 100% dans la zone 3 par rapport à la zone 1). L'impact de la végétation a été analysé à partir de la durée de fonte relative entre le début de la fonte et la disparition de la neige. Il apparait clairement que la végétation active la fonte précocement, allongeant ainsi significativement la durée de fonte (+600%). Cependant, l'impact de la végétation ne retarde pas la date de fin du couvert nival qui est de plus en plus précoce pour les zones 2 et 3. L'analyse spatiale à haute résolution montre que la présence arbustive entraine une date de fin de neige plus précoce par rapport au sol nu. Cette étude démontre clairement que la croissance de la végétation qui résulte du réchauffement climatique impacte la dynamique du couvert nival, aussi affectée par ce réchauffement. Une étude approfondie des processus en causes avec des mesures in situ appuyées par leur modélisation permettrait de mieux comprendre ces phénomènes. // Abstract : This study presents a spatial-temporal analysis of the joint evolution of the increase of shrubiness and the dynamics of snowmelt in Nunavik, Quebec, Canada. This zone is characterized by the complexity of the interaction of multiple changes of temperature, snow cover and vegetation growth. The first part of this study analyzes the temporal evolution of these changes. The analysis required the use of a large database on climate, satellite data and ground cover at multiple scales over a period of up to 60 years, from 1950 to 2012. The second part of the study consists of a spatial high-resolution analysis of the influence of the fraction of shrub cover on snowmelt. The analysis and interpretation of the results clearly show a significant climate change over the study area, divided into three latitudinal transects corresponding to tundra, open taiga and forested taiga. A significant warming of 0.75 ° C and 1.57 ° C per decade was experienced between zones 1 (tundra) and 3 (forested taiga) respectively. The maximum annual snow depth on the ground decreased over the 3 zones studied while winter precipitations increased in zones 1 and 3 over the last 45 years. The results show a significant increase in shrub vegetation in zones 2 and 3. The impact of the vegetation on snow was analyzed with melt duration (from melt onset to complete melt). It appears clearly that the vegetation triggers the melting process earlier and significantly extends the melt duration (+600%). However, the impact of vegetation does not delay the date of the snow cover disappearance. The high-resolution spatial analysis showed that shrubs cause an earlier snow cover disappearance date than bare soil. This study clearly demonstrates that vegetation growth resulting from global warming impacts the snow cover dynamics, which are also affected by global warming. A thorough study of the processes with in-situ measurements supported by models would help gaining a better comprehension of these phenomena.

Accurate knowledge of water content in seasonal snow can be helpful for water resource management. In this study, a distributed temperature index snow model based on temperature and precipitation as forcing data, is used to estimate snow storage in the Jollie catchment approximately 20km east of the main divide of the central Southern Alps, New Zealand. The main objective is to apply a frequently used assimilation method, the ensemble Kalman square root filter, to assimilate remotely sensed snow covered area into the model and evaluate the impacts of this approach on simulations of snow water equivalent. A 250m resolution remotely sensed data from Moderate Resolution Imaging Spectroradiometer (MODIS), specifically tuned to the study location was used. Temperature and precipitation were given on a 0.055 latitude/longitude grid. Precipitation was perturbed as input into the model, generating 100 ensemble members, which represented model error. Only observations of snow covered area that had less that 25% cloud cover classification were used in the assimilation precess. The error in the snow covered area observations was assumed to be 0.1 and grow linearly with cloud cover fraction up to 1 for a totally cloud covered pixel. As the model was not calibrated, two withholding experiments were conducted, in which observations withheld from the assimilation process were compared to the results. Two model states were updated in the assimilation, the total snow accumulation state variable and the total snow melt state variable. The results of this study indicate that the model underestimates snow storage at the end of winter and/or does not detect snow fall events during the ablation period. The assimilation method only affected simulated snow covered area and snow storage during the ablation period. That corresponded to higher correlation between modelled snow cover area and the updated state variables. Withholding experiments show good agreement between observations and simulated snow covered area. This study successfully applied the ensemble Kalman square root filter and showed its applicability for New Zealand conditions.

Les travaux présentés portent sur les étapes successives qui ont conduit au développement dans les années 1980 d'un modèle numérique qui simule l'évolution temporelle d'un manteau neigeux saisonnier en fonction des conditions météorologiques. Une première partie décrit le développement d'un modèle de neige multi-couches qui calcule les échanges d'énergie entre la neige et l'atmosphère et simule les principaux processus physiques qui contrôlent les échanges au sein du manteau neigeux.Une deuxième partie décrit comment ont été quantifiées les lois de métamorphose de la neige humide et de la neige sèche soumise à un faible gradient de température, de façon à compléter les connaissances existantes et proposer un jeu relativement complet de lois de métamorphoses de la neige saisonnière. Une troisième partie décrit l'implémentation de ces lois dans le modèle numérique, permettant ainsi de simuler la stratification du manteau neigeux, fonctionnalité qui n'existait dans aucun autre modèle à cette époque. Une évaluation détaillée de ce modèle sur le site du Col de Porte est présentée. La dernière partie introduit trois applications originales qui ont ensuite exploité les fonctionnalités de ce modèle : la simulation en temps réel de l'état caractéristique du manteau neigeux dans les Alpes françaises, l'étude de l'impact du changement climatique sur l'enneigement et la simulation de l'état du manteau neigeux dans un modèle hydrologique distribué / The thesis describes the different steps which lead during the 1980's to the development of an original numerical snow model. This model aimed at simulating the evolution of a seasonal snow cover as a function of the prevailing meteorological conditions. A first part describes the methods and algorithms used to compute the energy and mass exchange at the snow/atmosphere interface and inside the snowpack. The second part describes the experimental study which made possible the quantification of the metamorphism rate of snow samples submitted to weak temperature gradient and to liquid water content, in order to complete pre-existing knowledge on metamorphism. A third part describes the implementation of a set of metamorphism laws into the preliminary version of the snow model, which lead to the availability of the first numerical model able to simulate seasonal snowpack layering. The evaluation of the model at Col de Porte is presented. The last part introduces three applications of this model: real time monitoring of snowpack characteristics in the French Alps, assessment of the impact of climate change on snow climatology and simulation of the snowpack in a distributed hydrological model

This study is a result from a major investigation about the snow conditions in Sweden since the beginning of the twentieth century. For this purpose, data were analysed with respect to the maximum snow depth and the number of days with snow cover every year from some more than forty selected stations. These stations were then divided into different regions and means were calculated for each series. The data are presented in the shape of different histograms in the four following categories; the whole period in request (1900-2003), the latest 43 years (1961-2003), consecutive mean values for every decade and time series with the highest frequented fluctuations equalized. To be able to detect any trends in the plotted time series two statistical methods, simple linear regression and Mann-Kendall’s test, were applied. The calculations belonging to these tests are showed in tables. To be able to answer the question if the global warming can be related to the latest 3-4 decades predominantly warm winters in the southern part of Sweden I have been studying correlations in snow data with respect to the northern hemispheres mean temperature for the winter season. Corresponding estimates of the correlation coefficients have also been made with respect to the Swedish winter mean temperature. The response of the tests shows that it has not been such dramatic change in the snow conditions in the long run. The magnitude of the slope for the adjusted regression lines implies that the maximum snow depth and the number of days with snow cover in average have been on a fairly constant level during the latest hundred years. When it comes to the maximum snow depth one can distinguish a tendency for a small rise in Götaland and northern Norrland. This is also the only cases which are statistical significant for the period in request (1905-2003). For the shorter period 1961-2003 however, the number of days with snow cover has decreased quite substantially in the southern part of Sweden corresponding to a decrease about 40% in Götaland and 20% in Svealand. The test based on simple linear regression gives significant results in both cases while Mann-Kendall only establishes the trend for Götaland. A closer view of the maximum snow depth for the shorter period (1961-2003) does not give the same response but there is at least evidence for a significant decrease in Svealand in the test with simple linear regression. It corresponds to a decrease of about 30% since 1960. One cannot immediately relate the changes in the Swedish snow climate to the global warming. Estimated values of the correlation coefficient do not even give significant results for the period 1961-2003 despite of the fact that the global mean temperature has raised quite considerably since 1970. The corresponding calculations for the Swedish winter mean temperature show that it plays a very important roll if the precipitation in Götaland and Svealand is coming as rain or snow while it does not matter at all in northern Norrland. / Denna studie är ett resultat av en omfattande undersökning av snöförhållandena i Sverige sedan början av 1900-talet. Jag har för detta ändamål analyserat data av maximala snödjup och antalet dagar med snötäcke per kalenderår från ett 40-tal utvalda stationer. Dessa stationer har sedan delats upp på olika regioner varefter medelvärden har räknats fram i resp. fall. Datamaterialet illustreras här i form av olika stapeldiagram uppdelat på fyra följande kategorier; hela tidsserien, perioden 1961-2003, konsekutiva 10-årsmedelvärden samt en tidsserie med de mest högfrekventa svängningarna bortdämpade. För att kunna bedöma eventuella trender i de uppritade tidsserierna så har jag använt mig av de båda statistiska metoderna enkel linjär regression resp. Mann-Kendall's test. Tillhörande beräkningar redovisas på tabellform. För att svara på frågan om den globala uppvärmningen kan sättas i samband med de senaste 30-40 årens övervägande snöfattiga vintrar i södra Sverige så har jag studerat korrelationen av snödata gentemot det norra halvklotets vintermedeltemperatur. Motsvarande beräkningar av korrelationskoefficienter har också genomförts för den svenska vintermedeltemperaturen Utslaget på testerna visar att det inte har skett så dramatiska förändringar i snöförhållandena på lång sikt. Magnituden på lutningskoefficienten för de anpassade regressionslinjerna tyder på att det maximala snödjupet och antalet dagar med snötäcke i medeltal har legat på en ganska konstant nivå under de senaste hundra åren. När det gäller maximala snödjup så kan man paradoxalt nog se en tendens till en svag uppgång för Götaland och norra Norrland. Det är också de enda fallen som är statistiskt säkerställda för tidsserien som helhet. För den kortare perioden 1961-2003 så kan man däremot se att antalet dagar med snötäcke har minskat relativt kraftigt i södra Sverige motsvarande en nedgång på cirka 40% i Götaland och 20% i Svealand. Test med enkel linjär regression ger signifikanta resultat i båda fallen medan Mann-Kendall endast fastställer trenden för Götaland. En närmare undersökning av det maximala snödjupet för den kortare tidsserien ger dock inte lika tydligt utslag i statistiken men man kan trots allt urskilja en signifikant minskning för Svealand i testet med enkel linjär regression. Det rör sig här om en nedgång på cirka 30% efter 1960. Det går inte att omedelbart relatera förändringarna i det svenska snöklimatet till den globala uppvärmningen. Beräknade värden på korrelationskoefficienten ger inte ens signifikant utslag för perioden 1961-2003 trots att den globala medeltemperaturen har ökat ganska markant sedan 1970. Motsvarande beräkningar för den svenska vintermedeltemperaturen visar att den har väldigt stor betydelse för om nederbörden i Götaland och Svealand faller som regn eller snö medan det för norra Norrland inte har någon nämnvärd påverkan.

Effect of selected physical-geographical factors on the snow accumulation and snow melt Abstract: This master thesis analyzes the influence of physical-geographical factors on spatial distribution of snow water equivalent, and its evolution. In this work, emphasis is placed on describing the influence of vegetation, aspect and altitude. Measurement was carried out in experimental catchments Zlatý Brook and Bystřice River in western part of the Ore Mountains in winters 2008/2009, 2009/2010, 2010/2011 and 2011/2012. To evaluate the influence of these factors on value of snow water equivalent there was used one of the methods of multivariate statistical analysis - cluster analysis. The research shows that the greatest influence on the distribution and evolution of snow water equivalent in the experimental basins has vegetation and some dependency was proved also between the points of southern exposure. The measurement results demonstrate the suitability of cluster analysis for analyzing the data of point values of snow water equivalent. On the other hand the results showed the main limits of this method, especially the need for a large number of points with different characteristics. The results of measurements and statistical analysis are compared with results published in technical literature. Keywords: snow...