Uncertainty in Remote Sensing and GIS

The Effects of Uncertainty in Deposition Data on Predicting Exceedances of Acidity Critical Loads for Sensitive UK Ecosystems Part 1

Introduction Acidification can lead to harmful environmental effects, including depletion of fish stocks from lakes and streams, soil degradation and deforestation due to phytotoxi-city (Rodhe et al., 1995). The main sources of sulphur emissions, that in turn lead to sulphur deposition, are the burning of fossil fuels, for example, from coal-burning power stations and other […]

The Effects of Uncertainty in Deposition Data on Predicting Exceedances of Acidity Critical Loads for Sensitive UK Ecosystems Part 2

Monte Carlo uncertainty analysis The results of the Monte Carlo simulations are stochastic in nature so a result obtained from one run of the simulation will not necessarily be the same as from the next run. To check whether the number of calculations undertaken for the UK was sufficient to achieve reproducible results, the cumulative […]

Vertical and Horizontal Spatial Variation of Geostatistical Prediction Part 1

Introduction Spatially related data are found in many environmental applications. Measurements are generally available at a limited number of points only, and due to financial restrictions this number is often small. However, subsequent use of a variable often necessitates its availability for every point in the region of interest. For example, assume that yearly average […]

Vertical and Horizontal Spatial Variation of Geostatistical Prediction Part 2

Attempt to Relate Vertical and Horizontal Variation It is obvious from the previous sections that there is no simple analytic relation between horizontal and vertical variation. This is partly due to the fact that the vertical variability is locally restricted to one point in D whereas horizontal variation by definition relates to infinitely many points […]

Geostatistical Prediction and Simulation of the Lateral and Vertical Extent of Soil Horizons Part 1

Introduction Concerns about global climate change require estimates of various components of atmospheric and terrestrial carbon. This is particularly important with respect to terrestrial inorganic carbon, which is a major repository of global carbon dioxide (CO2). Global and regional estimates of inorganic carbon, particularly carbonates, are unreliable and in many cases do not take into […]

Geostatistical Prediction and Simulation of the Lateral and Vertical Extent of Soil Horizons Part 2

Methods and Analysis We stated earlier that a sensible approach to estimating local horizon thickness should involve two steps: an initial delineation of the zones where a horizon is most likely to be present and an optimal kriging of horizon thickness within these zones. Figure 14.1 Bourke irrigation area. Origin (bottom left): 337500 E, 6655000 […]

Geostatistical Prediction and Simulation of the Lateral and Vertical Extent of Soil Horizons Part 3

Results Comparing prediction bias using a simulated dataset The kriged transects (Figure 14.4b-e) are less variable than reality (Figure 14.4a). The transects and error histograms (Figure 14.5a-d) show that this results primarily from the overestimation of zero values, where no horizon is present, and to a lesser extent by the underestimation of large values. Of […]

Increasing the Accuracy of Predictions of Monthly Precipitation in Great Britain using Kriging with an External Drift Part 1

Introduction Maps of precipitation are used in a wide range of disciplines and in many different applications. Partly as a result of this, researchers have spent much effort to derive and test appropriate techniques for using data from rainfall monitoring networks to predict accurately precipitation at locations for which there are no observations. There is […]

Increasing the Accuracy of Predictions of Monthly Precipitation in Great Britain using Kriging with an External Drift Part 2

Analysis The results obtained using IDW, OK and KED were compared visually with maps (derived using a neighbourhood of 16 observations) and by cross-validation. With cross-validation, each observation is removed temporarily from the data set and predicted using the remaining data. The variograms were estimated and models were fitted using the Gstat package (Pebesma and […]

Conditional Simulation Applied to Uncertainty Assessment in DTMs Part 1

Introduction The computation of a digital terrain model (DTM) can be achieved using a stereoscopic pair of satellite sensor images. When an object is observed from two different view points, its projections on each image, which form a pair of homologous points, appear shifted and this shift, the disparity, is a known function of the […]

← Previous Entries

Next Entries →