Modelling and prediction of spatially distributed data such as the secondary cassiterite mineral distributions are often affected by spatial autocorrelation (SAC); a phenomenon that violates attributes data independence in space, which leads to type1 errors in classical statistics and overfitting or underfitting in machine learning (ML) classification respectively. The concept of overfitting and underfitting of spatially distributed datasets in an ML classification has not been properly addressed by the traditional random holdout technique of model validation, and this is a challenge to the assessment of predictive spatial model performance in spatially distributed datasets.

About 2,200 years ago a scholarly librarian in charge of the prestigious collections of the museum at Alexandria conducted an elegant exercise in logic and experimentation. Using seemingly unrelated bits of information such as the observation of the penetration of sunlight into a well in the city of Syene, the speed of a camel caravan, and the shadow cast by an obelisk in Alexandria, Eratosthenes calculated a remarkably accurate estimate of the circumference of the earth (Wilford 1981).

The integration of geology with data science disciplines, such as spatial statistics, remote sensing, and geographic information systems (GIS), has given rise to a shift in many natural sciences schools, pushing the boundaries of knowledge and enabling new discoveries in geological processes and earth systems.

When computers were frst invented in the middle of the last century, nobody ever anticipated that the instruction sets to program them would ultimately evolve to the point where many of them could be accessed freely, without charge, in ways that would become widely available to nonexpert communities. Nor did we ever consider that computers would penetrate every corner of social life, revolutionizing our social behavior, our science, and our politics.

What is cartography today? When I look back on the first and second editions of this book, it’s almost like seeing cartography as it progressed since 2009 and it now looks almost nothing like it did then. The expansion of what’s possible has been incredible. Massive technological change, accompanied by free, open, highly detailed, and reliable datasets, along with a surge of interest in the field from all corners, combined to make what was once impossible cartographically, possible.

QGIS is a crossing point of the free and open source geospatial world. While there are a great many tools in QGIS, it is not one massive application that does everything, and it was never really designed to be that from the beginning. It is rather a visual interface to much of the open source geospatial world. You can load data from proprietary and open formats into spatial databases of various flavors and then analyze the data with well-known analytical backends before creating a printed or web-based map to display and interact with your results. What’s QGIS’s role in all this? It’s the place where you check your data along the way, build and queue the analysis, visualize the results, and develop cartographic end products. This learning path will teach you all that and more, in a hands-on learn-by-doing manner. Become an expert in QGIS with this useful companion. <...>

Geography has always been important to humans. Stone-age hunters anticipated the location of their quarry, early explorers lived or died by their knowledge of geography, and current societies work and play based on their understanding of who belongs where. Applied geography, in the form of maps and spatial information, has served discovery, planning, cooperation, and conflict for at least the past 3000 years (Figure 1-1). Maps are among the most beautiful and useful documents of human civilization.

The natural resources on the earth seem to be randomly distributed but their variations over space and time are not all random. They exhibit a spatial correlation. This spatial correlation can be captured by geostatistics. Geostatistics deals with the analysis and modelling of geo-referenced data. The point observations are analyzed and interpolated to create spatial maps. For geostatistical interpolation, first the spatial correlation structures of the parameter of interest are quantified and then spatial interpolation is done using the quantified spatial correlation and optimal predictions at unobserved locations to create a map.

Вы приступаете к изучению расширения к ArcGIS компании ESRI®  модуля Geostatistical Analyst, предназначенного для усовершенствованного моделирова ния поверхности с использованием детерминистских и геостатистических методов. Модуль Geostatistical Analyst расширяет возможности ArcMap за счет появления дополнительных инструментов, предназначенных для исследовательского анализа пространственных данных, а также Мастера операций геостатистики, который поможет вам в процессе построения статистически достоверной поверхности. Поверхности, создаваемые с помощью модуля Geostatistical Analyst, могут быть впоследствии использованы в моделях ГИС и для визуализации, в том числе с использованием таких расширений ArcGIS, как ArcGIS Spatial Analyst и 3D Analyst. <...>