Filtering by
- All Subjects: Landscape changes
- Creators: Cerveny, Randall S.
Full simulations originating in climate modeling have been the conventional approach to impacts assessment. But, once debatable climate projections are applied to hydrologic models challenged to accurately represent the region’s arid hydrology, the range of possible scenarios enlarges as uncertainties propagate through sequential levels of modeling complexity. Numerous issues render future projections frustratingly uncertain, leading many researchers to conclude it will be some decades before hydroclimatic modeling can provide specific and useful information to water management.
Alternatively, this research investigation inverts the standard approach to vulnerability assessment and begins with characterization of the threatened system, proceeding backwards to the uncertain climate future. Thorough statistical analysis of historical watershed climate and runoff enabled development of (a) a stochastic simulation methodology for net basin supply (NBS) that renders the entire range of droughts, and (b) hydrologic sensitivities to temperature and precipitation changes. An operations simulation model was developed for assessing the SRP reservoir system’s cumulative response to inflow variability and change. After analysis of the current system’s drought response, a set of climate change forecasts for the balance of this century were developed and translated through hydrologic sensitivities to drive alternative NBS time series assessed by reservoir operations modeling.
Statistically significant changes in key metrics were found for climate change forecasts, but the risk of reservoir depletion was found to remain zero. System outcomes fall within ranges to which water management is capable of responding. Actions taken to address natural variability are likely to be the same considered for climate change adaptation. This research approach provides specific risk assessments per unambiguous methods grounded in observational evidence in contrast to the uncertain projections thus far prepared for the region.
Remote sensing has demonstrated to be an instrumental tool in monitoring land changes as a result of anthropogenic change or natural disasters. Most disaster studies have focused on large-scale events with few analyzing small-scale disasters such as tornadoes. These studies have only provided a damage assessment perspective with the continued need to assess reconstruction. This study attempts to fill that void by examining recovery from the 1999 Moore, Oklahoma Tornado utilizing Landsat TM and ETM+ imagery. Recovery was assessed for 2000, 2001 and 2002 using spectral enhancements (vegetative and urban indices and a combination of the two), a recovery index and different statistical thresholds. Classification accuracy assessments were performed to determine the precision of recovery and select the best results. This analysis proved that medium resolution imagery could be used in conjunction with geospatial techniques to capture recovery. The new indices, Shortwave Infrared Index (SWIRI) and Coupled Vegetation and Urban Index (CVUI), developed for disaster management, were the most effective at discerning reconstruction using the 1.5 standard deviation threshold. Recovery rates for F-scale damages revealed that the most incredibly damaged areas associated with an F5 rating were the slowest to recover, while the lesser damaged areas associated with F1-F3 ratings were the quickest to rebuild. These findings were consistent for 2000, 2001 and 2002 also exposing that complete recovery was never attained in any of the F-scale damage zones by 2002. This study illustrates the significance the biophysical impact has on recovery as well as the effectiveness of using medium resolution imagery such as Landsat in future research.