Summer Summary

This post is a summary of the main ideas collected during the research period of summer. In total, I read completely 16 articles, and another number of incompletely readings. The research focused mainly in two topics: my present research, which is stormwater sustainability, and optimization techniques, mainly focused in multi-objective genetic algorithm.

For the first and more important topic of my research in this summer, the readings helped me to support the main ideas of the hydrologic footprint residence. HFR is a new metric used to assess the impacts of urbanization in the hydrologic flow regime. Many articles served as references to characterize the impact of urbanization in the hydrological regime. The transformation of natural land covers as forest and grassland to impervious covers alters most of the hydrological processes such as interception, evapotranspiration, infiltration and overland runoff.

In the revised literature, it was identified the main hydrologic metrics used to assess the impact of urbanization. Most of the metrics are based in time series of flow discharges. The main metrics found are: flow frequency curve, flow duration curve, TQmean, which is fraction of year that flow rate exceeds the mean flow rate of the entire flow time series, TQ0.5yr, which is the fraction of year that the flow rate exceeds the historical half-year peak flow rate, peak flow, storm flow duration, time-to-peak, ratio of time-to-peak to storm flow event duration, event-mean rise rate, event-mean fall rate, duration that the flow rate exceeds 75% of the peak flow rate in an event, ratio of event-peak-duration to total event duration, and IHA (indices of hydrological alteration). Those metrics can be used to evaluate different alternatives of development, Best Management Practices (BMPs) and Low Impact Development (LID).

Besides that, an interesting discussion about the effectiveness of BMPs was addressed in some articles. The argument is BMPs introduces a sustained flow for most of the storms which can harm downstream communities. BMPs are design to attenuate bigger storms. Besides that, urbanization increases the frequency of flow events. The combination of more frequent events with sustained flow is being questioned by recent researches.

Another topics explored during this research are the links between the hydrological flow regime and heath of instream ecosystems. Poff et al (1997) summarizes the knowledge about formation of habitats and the natural hydrological regime. Scientific community agrees that stream is a master variable for instream environments because it regulates most of biological and physicochemical processes in rivers. The natural variability of low and high flows plays a crucial role in the entire life cycle of instream and riverine species. Watershed management has been limited to water quality and one aspect of water quantity, which is minimum flow.

The second set of articles is related to optimization algorithms. Two interesting articles present a new heuristic optimization technique called Harmony Search (HS). HS is inspired the way musician improvise harmony searching for maximize the aesthetic estimation of music (objective function). The decision variables are the pitches (sounds) musicians play. Several problems are tested in the articles, and HS presented better performance when compared to other optimization techniques such as Generalized Reduced Gradient (GRC), Evolutionary Programming (EP), Genetic Algorithm (GA), and Nonlinear Programming Gradient (NLPG). This better efficiency, however, should be looked with attention, since performance of methods can be highly dependent of parameters and the problems itself.

The last group of articles is focused in Multi-objective optimization techniques, more specifically genetic algorithms. The objective here is to understand the main operators used in different techniques. Two methods have been studied in more details: MOGA and NSGA-II. One article combined the use of MOGA and Constrained Differential Dynamic Programming in a surface reservoir allocation problem.

In summary, important goals have been achieved during this summer period. The idea of reading two papers a week, one specific of my research and other more general, to discuss in the research group is really efficient. My reading and writing skills have improved a lot in period. I hope we can continue to do that with, hopefully, more students contributing to the discussion.