Browsing University of Alaska Fairbanks by Subject "forecasting"
Now showing items 1-3 of 3
An evaluation of the use of moderate resolution imaging spectroradiometer (MODIS)-derived aerosol optical depth to estimate ground level PM2.5 in AlaskaThe air quality monitoring (AQM) network in Alaska is limited to major urban areas and national parks thus leaving a large proportion of the state unmonitored. To evaluate the use of Moderate Resolution Imaging Spectroradiometer (MODIS)-derived aerosol optical depth (AOD) to predict ground-level PM2.5 concentrations and thereby increase the spatial coverage of the AQM network in Alaska, MODIS AOD was first validated against ground-based measurements of AOD in Utqiagvik and Bonanza Creek Alaska. MODIS AOD from 2000 to 2014 was obtained from MODIS collection 6 using the dark target land and ocean algorithms between the months of April and October. Based on validation results, individual Aqua and Terra products are valid for both locations at 10-kilometer and 3-kilometer resolution. In addition, combined Aqua and Terra MODIS AOD products are valid for both locations at 3-kilometer resolution and 10-kilometer resolution for Utqiagvik. The available PM2.5 data was then compared for satellite retrieval and all retrieval days to determine if there was sufficient data and the amount of bias introduced by possible low retrieval rates. Overall, Juneau had the lowest retrieval rates while Fairbanks and North Pole had the highest retrieval rates. In addition, Juneau appeared to have relatively high bias while stations located in Anchorage, Palmer, Fairbanks and North Pole had relatively low bias. Based on these findings, no models were developed for Juneau (southeast Alaska). Multilinear regression models were then developed for southcentral (Anchorage and Palmer) and interior (Fairbanks and North Pole) Alaska where the log-transform of PM2.5 was the response and meteorological data and the log-transform of MODIS AOD were the predictors. MODIS AOD appeared to be most highly correlated with PM2.5 in interior Alaska, while there was little to no correlation between MODIS AOD and PM2.5 in southcentral Alaska. All models underestimate surface PM2.5 concentrations which may be due to the high percentage of low PM2.5 values used to develop the models and the limited retrieval rates. Alternative modeling methods such as mixed-effects modeling may be necessary to develop adequate models for predicting surface PM2.5 concentrations. The MLR models did not perform well and should not be used to predict ground-based PM2.5 concentrations. Further research using alternative modeling methods should be performed. Model performance may also be improved by only using higher concentrations of PM2.5 to develop models. Overall, the limited spatial coverage of Alaska's air quality monitoring network and the low temporal resolution of MODIS-derived AOD make modeling the relationship between MODIS AOD and PM2.5 difficult in Alaska.
Probabilistic decline curve analysis in unconventional reservoirs using Bayesian and approximate Bayesian inferenceIn this work, a probabilistic methodology for Decline Curve Analysis (DCA) in unconventional reservoirs is presented using a combination of Bayesian statistical methods and deterministic models. Accurate reserve estimation and uncertainty quantification are the primary objectives of this study. The Bayesian inferencing techniques described in this work utilizes three sampling mechanisms, namely the Gibbs Sampling (implemented in OpenBUGS), the Metropolis Algorithm, and Approximate Bayesian Computation (ABC) to sample parameter values from their posterior distributions. These different sampling mechanisms are applied in conjunction with DCA models like Arps, Power Law Exponential (PLE), Stretched Exponential Production Decline (SEPD), Duong and Logistic Growth Analysis (LGA) to estimate prediction intervals. Production is forecasted, and uncertainty bounds are established using these prediction intervals. A complete workflow and the summary steps for each of the sampling techniques are provided to permit readers to replicate results. To examine the reliability, the methodology was tested over 74 oil and gas wells located in the three main sub plays of the Permian Basin, namely, the Delaware play, the Central Basin Platform, and the Midland play. Results show that the examined DCA-Bayesian models are successful in providing a high coverage rate, low production prediction errors and narrow uncertainty bounds for the production history data sets. The methodology was also successfully applied to unconventional reservoirs with as low as 6 months of available production history. Depending on the amount of production history available, the combined deterministic-stochastic model that provides the best fit can vary. It is therefore recommended that all possible combinations of the deterministic and stochastic models be applied to the available production history data. This is in order to obtain more confidence in the conclusions related to the reserve estimates and uncertainty bounds. The novelty of this methodology relies in using multiple combinations of DCA-Bayesian models to achieve accurate reserve estimates and narrow uncertainty bounds. The paper can help assess shale plays as most of the shale plays are in the early stages of production when the reserve estimations are carried out.