Predictive Mapping of Reef Fish Species Richness, Diversity and Biomass in Zanzibar using IKONOS Imagery and Machine-learning Techniques

Remote Sensing of Environment, Volume 114, Issue 6, 15 June 2010, Pages 1230-1241

Anders Knudby, Ellsworth LeDrew, and Alexander Brenning

“During the last three decades, the large spatial coverage of remote sensing data has been used in coral reef research to map dominant substrate types, geomorphologic zones, and bathymetry. During the same period, field studies have documented statistical relationships between variables quantifying aspects of the reef habitat and its fish community. Although the results of these studies are ambiguous, some habitat variables have frequently been found to correlate with one or more aspects of the fish community. Several of these habitat variables, including depth, the structural complexity of the substrate, and live coral cover, are possible to estimate with remote sensing data. In this study, we combine a set of statistical and machine-learning models with habitat variables derived from IKONOS data to produce spatially explicit predictions of the species richness, biomass, and diversity of the fish community around two reefs in Zanzibar. In the process, we assess the ability of IKONOS imagery to estimate live coral cover, structural complexity and habitat diversity, and we explore the importance of habitat variables, at a range of spatial scales, in the predictive models using a permutation-based technique. Our findings indicate that structural complexity at a fine spatial scale (not,  vert, similar 5 to 10 m) is the most important habitat variable in predictive models of fish species richness and diversity, whereas other variables such as depth, habitat diversity, and structural complexity at coarser spatial scales contribute to predictions of biomass. In addition, our results demonstrate that complex model types such as tree-based ensemble techniques provide superior predictive performance compared to the more frequently used linear models, achieving a reduction of the cross-validated root-mean-squared prediction error of 3–11%. Although aerial photographs and airborne lidar instruments have recently been used to produce spatially explicit predictions of reef fish community variables, our study illustrates the possibility of doing so with satellite data. The ability to use satellite data may bring the cost of creating such maps within the reach of both spatial ecology researchers and the wide range of organizations involved in marine spatial planning.”

Spatial Heterogeneity in the Shrub Tundra Ecotone in the Mackenzie Delta Region, Northwest Territories: Implications for Arctic Environmental Change

Ecosystems, Volume 13, Number 2 / March, 2010

Trevor C. Lantz, Sarah E. Gergel and Steven V. Kokelj

“Growing evidence suggests that plant communities in the Low Arctic are responding to recent increases in air temperature. Changes to vegetation, particularly shifts in the abundance of upright shrubs, can influence surface energy balance (albedo), sensible and latent heat flux (evapotranspiration), snow conditions, and the ground thermal regime. Understanding fine-scale variability in vegetation across the shrub tundra ecotone is therefore essential as a monitoring baseline. In this article, we use object-based classifications of airphotos to examine changes in vegetation characteristics (cover and patch size) across a latitudinal gradient in the Mackenzie Delta uplands. This area is frequently mapped as homogenous vegetation, but it exhibits fine-scale variability in cover and patch size. Our results show that the total area and size of individual patches of shrub tundra decrease with increasing latitude. The gradual nature of this transition and its correlation with latitudinal variation in temperature suggests that the position of the shrub ecotone will be sensitive to continued warming. The impacts of vegetation structure on ecological processes make improved understanding of this heterogeneity critical to biophysical models of Low Arctic ecosystems.”

A Fuzzy Set Based Approach for Integration of Thematic Maps for Landslide Susceptibility Zonation

Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, Volume 3, Issue 1, 2009, Pages 30 – 43

D. P. Kanungo, M. K. Arora, S. Sarkar, and R. P. Gupta

“Spatial prediction of landslides is termed landslide susceptibility zonation (LSZ). In this study, an objective weighting approach based on fuzzy concepts is used for LSZ in a part of the Darjeeling Himalayas. Relevant thematic layers pertaining to landslide causative factors have been generated using remote sensing and geographic information system (GIS) techniques. The membership values for each category of thematic layers have been determined using the cosine amplitude fuzzy similarity method and are used as ratings. The integration of these ratings led to the generation of LSZ map. The integration of different ratings to generate an LSZ map has been performed using a fuzzy gamma operator apart from the arithmetic overlay approach. The process is based on determination of combined rating known as the landslide susceptibility index (LSI) for all the pixels using the fuzzy gamma operator and classification using the success rate curve method to prepare the LSZ map. The results indicate that as the gamma value increases, the accuracy of the LSZ map also increases. It is observed that the LSZ map produced by the fuzzy algebraic sum has reflected a more real situation in terms of landslides in the study area.”

Learn to Make and Share High-Quality Maps Faster than Ever

Log In to an ESRI Live Training Seminar to Get an Overview of the New Features in ArcGIS 10 ArcMap

To introduce users to what’s new in ArcMap, ESRI will host the live training seminar Using ArcMap in ArcGIS Desktop 10. It will air at www.esri.com/lts on Thursday, April 29, 2010, at 9:00 a.m., 11:00 a.m., and 3:00 p.m. Pacific daylight time.

The seminar will introduce new features in ArcMap that will reduce time spent on common mapping tasks and improve the quality of map products. The instructor will demonstrate how to use these features to speed up workflows.

Attendees will learn about

  • Improvements to the ArcMap interface, including a new configurable, embedded Catalog window that allows users to customize and maximize their map view
  • New features to help users work faster, including an instant search tool and access to professionally designed online basemaps
  • Improved cartographic capabilities that enable data-driven queries to search for and highlight specific map features, streamlining the creation of custom maps from a map series, which can then be exported as PDFs
  • Additional resources for learning more about new features in ArcGIS 10

This live training seminar will help cartographers and other geographic information system (GIS) professionals who make maps understand how ArcMap in ArcGIS 10 will improve their workflow and productivity. A broadband Internet connection and an ESRI Global Account are needed to participate in the training seminar. Creating a global account is easy and free: visit www.esri.com/lts, click Login, and register your name and address. A few weeks after the live presentation, this seminar will be archived and available for viewing on the ESRI Training Web site.

[Source: ESRI press release]

Detecting Negative Spatial Autocorrelation in Georeferenced Random Variables

International Journal of Geographical Information Science, Volume 24, Issue 3 March 2010 , pages 417 – 437

Daniel A. Griffith; Giuseppe Arbia

“Negative spatial autocorrelation refers to a geographic distribution of values, or a map pattern, in which the neighbors of locations with large values have small values, the neighbors of locations with intermediate values have intermediate values, and the neighbors of locations with small values have large values. Little is known about negative spatial autocorrelation and its consequences in statistical inference in general, and regression-based inference in particular, with spatial researchers to date concentrating mostly on understanding the much more frequently encountered case of positive spatial autocorrelation. What are the spatial contexts within which negative spatial autocorrelation should be readily found? What are its inferential consequences for regression models? This paper presents selected empirical examples of negative spatial autocorrelation, adding to the slowly growing literature about this phenomenon.”

Space–Time Geostatistics for Geography: A Case Study of Radiation Monitoring Across Parts of Germany

Geographical Analysis, Volume 42 Issue 2, Pages 161 – 179, Published Online 13 Apr 2010

Gerard B. M. Heuvelink and Daniel A. Griffith

“Many branches within geography deal with variables that vary not only in space but also in time. Therefore, conventional geostatistics needs to be extended with methods that estimate and quantify spatiotemporal variation and use it in spatiotemporal interpolation and stochastic simulation. This article briefly summarizes the main concepts of space–time geostatistics. Kriging in space and time can be done in much the same way as it is in a purely spatial setting. The main difficulties are in defining a realistic stochastic model that is assumed to have generated data and in characterizing and estimating the space–time correlation of that model. This article uses a model-based geostatistical approach to characterize space–time variability. The space–time variable of interest is treated as a sum of independent stationary spatial, temporal, and spatiotemporal components, which leads to a sum-metric space–time variogram model. Methods are illustrated with a case study of space–time interpolation of monthly averages of detected background radiation for a 5-year period in four German states.”

A New GIS Nitrogen Trading Tool Concept for Conservation and Reduction of Reactive Nitrogen Losses to the Environment

Advances in Agronomy, Volume 105, 2010, Chapter Chapter 4, Pages 117-171

J.A. Delgado, C.M. Gross, H. Lal, H. Cover, P. Gagliardi, S.P. McKinney, E. Hesketh, and M.J. Shaffer

“Nitrogen (N) inputs to agricultural systems are important for their sustainability. However, when N inputs are unnecessarily high, the excess can contribute to greater agricultural N losses that impact air, surface water, and groundwater quality. It is paramount to reduce off-site transport of N by using sound management practices. These practices could potentially be integrated with water and air quality markets, and new tools will be necessary to calculate potential nitrogen savings available for trade. The USDA-NRCS and USDA-ARS Soil Plant Nutrient Research Unit developed a web-based and stand-alone Nitrogen Trading Tool (NTT) prototype. These prototypes have an easy-to-use interface where nitrogen management practices are selected for a given state and the NTT calculates the nitrogen trading potential compared to a given baseline. The stand-alone prototype can also be used to calculate potential savings in direct and indirect carbon sequestration equivalents from practices that reduce N losses. These tools are powerful, versatile, and can run with the USA soil databases from NRCS (SSURGO) and NRCS climate databases. The NTT uses the NLEAP model, which is accurate at the field level and has GIS capabilities. Results indicate that the NTT was able to evaluate management practices for Ohio, Colorado, and Virginia, and that it could be used to quickly conduct assessments of nitrogen savings that can potentially be traded for direct and indirect carbon sequestration equivalents in national and international water and air quality markets. These prototypes could facilitate determining ideal areas to implement management practices that will mitigate N losses in hot spots and provide benefits in trading.”