The operation of Glen Canyon Dam on the Colorado River affects several downstream resources and water uses including water supply for consumptive uses in Arizona, California, and Nevada, hydroelectric power production, endangered species of native fish, recreational angling for non-native fish, and recreational boating in the Grand Canyon. Decisions about the magnitude and timing of water releases through the dam involve trade-offs between these resources and uses. The numerous laws affecting dam operations create a hierarchy of legal priorities that should govern these decisions. At the top of the hierarchy are mandatory requirements for water storage and delivery and for conservation of endangered species. Other resources and water uses have lower legal priorities. The Glen Canyon Dam Adaptive Management Program ("AMP") has substituted collaborative decision making among stakeholders for the hierarchy of priorities created by law. The AMP has thereby facilitated non-compliance with the Endangered Species Act by the Bureau of Reclamation, which operates the dam, and has effectively given hydroelectric power production and non-native fisheries higher priorities than they are legally entitled to. Adaptive management is consistent with the laws governing operation of Glen Canyon Dam, but collaborative decision making is not. Nor is collaborative decision making an essential, or even logical, component of adaptive management. As implemented in the case of Glen Canyon Dam, collaborative decision making has actually stifled adaptive management by making agreement among stakeholders a prerequisite to changes in the operation of the dam. This Article proposes a program for adaptive, but not collaborative, management of Glen Canyon Dam that would better conform to the law and would be more amenable to adaptation and experimentation than would the current, stakeholder-centered program.

The purpose of this project is to create a useful tool for musicians that utilizes the harmonic content of their playing to recommend new, relevant chords to play. This is done by training various Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) on the lead sheets of 100 different jazz standards. A total of 200 unique datasets were produced and tested, resulting in the prediction of nearly 51 million chords. A note-prediction accuracy of 82.1% and a chord-prediction accuracy of 34.5% were achieved across all datasets. Methods of data representation that were rooted in valid music theory frameworks were found to increase the efficacy of harmonic prediction by up to 6%. Optimal LSTM input sizes were also determined for each method of data representation.