Value Chains And Resilient Coastal Communities In The Nordic Atlantic PDF Download

Global supply chain is largely dependent on seaports and marine terminals. Ports serve international cargo traffic for imports and exports. About 90% of the world’s goods are transported by sea and over 70% as containerized cargo ships. Coastal hurricanes/cyclones and rainfall flood disasters cause major disruptions for sea shipping traffic, disruptions of port infrastructure, and adverse impacts on coastal communities each year. Additionally, these weather related disasters threaten millions of people, damage infrastructure, and cost billions of dollars in global supply chain disruptions. Sustainable global supply chains, port infrastructure, and coastal community impact by these extreme weather disasters are the major motivation of this research. The objectives of this research are: (1) modeling shipping demand and level of service, (2) developing Landsat-8 satellite imagery based methodology for mapping surface types and landuse, and (3) assessing the impact of coastal disasters and climate related sea level rise. The Autoregressive Integrated Moving Average (ARIMA) model equations, the Artificial Neural Networks (ANN) models, and regression equations were developed using historical containerized cargo volumes to predict the future volumes for the Port of New Orleans and the Port of New York and New Jersey. The predictions by these models indicate that the ANN model achieves the most accurate predicted values, compared to reported volume. However, the ANN approach requires future values of independent variable inputs to calculate the forecast. Therefore, applying the ANN model was recommended for short-term prediction for these ports. The ARIMA model equation was applied for long-term prediction because it does not need other independent variable inputs. Results of cargo vessel volume analysis for ten selected international shipping navigation routes using Automatic Identification System (AIS) data show that the Europe Atlantic route to the East Coast of the U.S. has the largest cargo vessel volumes. A spatial map of cargo vessel demand for selected navigation routes was also created. A level of service (LOS) methodology for cargo vessel service was developed using AIS data for the Port of Miami to evaluate the operating conditions of a seaport. A mathematical function to estimate LOS level (A, B, C, D, E, F) was proposed based on delay time and waiting time of cargo vessels at the port and number of processed cargo vessels per total annual cargo vessels. A new methodology was developed to classify built and non-built surfaces using Landsat-8 satellite imagery. Groundtruth samples of the Landsat-8 pansharpened multispectral satellite imageries from six selected sites were sampled and used to develop the Landsat-8 Built-up Area and Natural Surface (L-BANS) auto-classification methodology. The L-BANS surface classification results for most sites using GeoMedia Pro geospatial analysis were within ±15% of the groundtruth. Based on analysis of variance (ANOVA) hypothesis testing results, the difference between the L-BANS results and the groundtruth was not statistically significant. The anthropogenic CO2 based global warming hypothesis was evaluated to undertand climate impacts. Measured global temperature and atmospheric carbon dioxide (CO2) data from 1958 to 2016 were analyzed. The final ARIMA time series seasonal model equation for monthly global temperature data had a high R value of 0.989 with only 2.25% difference compared to measured values. The final ARIMA model equation for monthly CO2 data provided reasonably accurate results for 2016 monthly measured CO2 data with high a R value of 0.999 with only 0.0025% difference compared to measured values. The results show that there is very poor crosscorrelation (0.08) between global temperature and CO2. Both IPCC and EPA models predict unreasonably high values of CO2 until 2050. This research shows that contrary to the IPCC claims, global warming is not caused by anthropogenic CO2. (Abstract shortened by ProQuest.)