Program development plan : automation of field operations and services /

On spine: AFOS program development plan.

Port promotion, protection and administration; a Baltimore Association of Commerce Staff Field survey and report on eight Atlantic and Gulf ports, March-May, 195

Mode of access: Internet.

Annual Report of Target Practice with Small Arms and 3-Inch Field Pieces and Boat Guns

Mode of access: Internet.

Essential elements and soil amendments for plants : sources and use for agriculture /

Mode of access: Internet.

Transactions of the Edinburgh Field Naturalists' and Microscopical Society

"Institutes as the Edinburgh naturalists' field club."

A treatise on rocks, rock-weathering and soils /

Includes bibliographical references and index.

The great world's farm; some account of nature's crops and how they are grown,

2d ed.

Response of the salt-freshwater interface in a coastal aquifer to a wave-induced groundwater pulse: field observations and modelling

Field observations on an unconfined coastal aquifer showed that a groundwater pulse, generated by it moderate (significant wave height, H-sig similar to 4.5 m) wave/storm event, induced significant oscillations in the salt-freshwater interface of the order of several metres in the horizontal direction. A dynamic sharp-interface model is developed to quantify the mechanism of these interface oscillations. The model uses the 50% seawater salinity contour as the location of the equivalent sharp-interface. The model was calibrated against the observed groundwater table fluctuations. It predicted reasonably well the interface oscillations with a slight over-prediction of the oscillation magnitude and a steepening of the interface. The neglect of mixing in the salt-freshwater mixing zone by the sharp-interface model is suggested as a possible contributor to the discrepancies between the model predictions and observations. In contrast with the significant wave effects, there was no observable response of the interface to diurnal or semidiurnal tides. (C) 2004 Elsevier Ltd. All rights reserved.

Cattle, crops and clearing: Regional drivers of landscape change in the Brigalow Belt, Queensland, Australia, 1840-2004

Landscape change occurs through the interaction of a multitude of natural and human driving forces at a range of organisational levels, with humans playing an increasingly dominant role in many regions of the world. Building on the current knowledge of the underlying drivers of landscape change, a conceptual framework of regional landscape change was developed which integrated population, economic and cultural values, policy and science/technology. Using the Southern Brigalow Belt biogeographic region of Queensland as a case study, the role of natural and human drivers in landscape change was investigated in four phases of settlement since 1840. The Brigalow Belt has experienced comparable rates of vegetation clearance over the past 50 years to areas of tropical deforestation. Economic factors were important during all phases of development, but the five regional drivers often acted in synergy. Environmental constraints played a significant role in slowing rates of change. Temporal trends of deforestation followed a sigmoidal curve, with initial slow change accelerating though the middle phases then slowing in recent times. Future landscape management needs to take account of the influence of all the components of the conceptual framework, at a range of organisational levels, if more ecologically sustainable outcomes are to be achieved. (c) 2006 Elsevier B.V. All rights reserved.

Random field models and priors on wind

This report seeks to make concrete some of the ideas we have been discussing about sensible priors for winds over the ocean. In particular, random field models are reviewed, as are permissible covariance functions. The criteria which these covariance functions must satisfy in order that vorticity and divergence exist and are continuous are defined. The use of Helmholtz theorem is discussed, and possible choices for the covariances are suggested.

Gaussian processes and SVM: Mean field results and leave-one-out

In this chapter, we elaborate on the well-known relationship between Gaussian processes (GP) and Support Vector Machines (SVM). Secondly, we present approximate solutions for two computational problems arising in GP and SVM. The first one is the calculation of the posterior mean for GP classifiers using a `naive' mean field approach. The second one is a leave-one-out estimator for the generalization error of SVM based on a linear response method. Simulation results on a benchmark dataset show similar performances for the GP mean field algorithm and the SVM algorithm. The approximate leave-one-out estimator is found to be in very good agreement with the exact leave-one-out error.