North American Summer Monsoon: Understanding Its Interannual and Intra-annual Variability and Implications to Water Resources Management
Presiding: B Rajagopalan, University of Colorado; A Barros, Duke University
H51F-01 INVITED 08:30h
The North American Monsoon Experiment: An integrated hydroclimatic research program for southwestern North America
The North American Monsoon Experiment (NAME) is a multi-year, multi-agency, international research program designed to improve the understanding and prediction of the warm season climate over southwestern North America. The multi-tiered structure of the NAME program is designed to highlight cross-scale interactions which modulate the regional hydroclimatic system. During the summer of 2004 an Enhanced Observation Period (EOP) was conducted in which strategic augmentations to regional land surface, oceanic and atmospheric observing networks were made. Measurements were concentrated in and around the Gulf of California, western Mexico and across the southwestern U.S. Data from the summer's activities have already shed light on a number of key regional scale processes controlling both variability and persistence characteristics of precipitation. Along with an overview of findings from the 2004 EOP, core elements of the NAME research program will be presented. Current and ongoing research in precipitation and streamflow interactions within the NAME region will be emphasized. The presentation will conclude with a discussion on plans for integrating hydrological and climate science research within the NAME program.
H51F-02 08:45h
Weather and Climate Prediction for the North American Monsoon
Some of the major elements of the North American monsoon include the onset and seasonal behavior of precipitation, the moisture sources, orographic responses, effects of sea surface temperature (SST) anomalies over the Gulf of Mexico, Pacific and Atlantic Oceans, and the teleconnection with the intertropical convergence zone (ITCZ). This study addresses these issues on the medium range (a week) to seasonal (3 month) time scales. Our approach is one of constructing ensemble forecasts that include 11 weather models for the medium range and 13 coupled atmosphere-ocean models for seasonal time scales. The metrics for forecasts evaluation include deterministic measures such as RMS error and anomaly correlation, and probabilistic measures such as the equitable threat scores and Briar skill scores. The ensemble forecast approach includes a conventional FSU superensemble for weather and a variant called the synthetic superensemble for the seasonal climate. These superensemble strings covering a 13-year period show that it is possible to predict some of the important features of the North American monsoon at a higher skill with the superensemble compared to the participating member models.
H51F-03 09:00h
The Role of Convection in the North American Monsoon
The North American Monsoon (NAM) is a very important part of the hydrologic cycle of the southwestern US and northwest Mexico. Improving the understanding and prediction of the NAM system is crucial for efficient management of regional water resources. We use the Regional Atmospheric Modeling System (RAMS) with different configurations to simulate the evolution of the NAM and explore the role played by convection in that process. We employ a high spatial resolution and ADAP, a new vertical coordinate system for a better representation of the complex terrain associated with the Sierra Madre mountains and the convective clouds and precipitation they spawn. ADAP is a form of shaved grid cell method that uses the finite volume instead of the finite difference scheme used in the traditional terrain-following coordinate systems in other models. The results are validated against satellite and raingauge data. We find that properly resolving convection, instead of relying on parameterizations, is the key to improving the simulation of the NAM system.
H51F-04 09:15h
Interannual Variability of Summertime Rainfall Over the Southwestern US: How Much of it is Climate-related and how Much is Related to Chance?
Many investigations of the North American monsoon system, particularly as it impacts northwestern Mexico and the southwestern United States, have focused upon summertime precipitation and its interannual variations. Here we study the interannual variance of summertime precipitation at 78 stations in the southwestern US using daily Markov Chain models and empirical intensity distributions. Modeling results suggest that a second-order, daily Markov Chain model with stationary (i.e. non-varying) interannual event frequency and intensity characteristics can capture over 75% of the interannual variance in the seasonal number of wet days in the region and 85% of the interannual variance in the total summertime precipitation. These results indicate that a large fraction of the interannual variance in seasonal precipitation can be explained simply by the random evolution of daily rainfall within the season itself, making it inherently difficult to predict. In addition, only a small fraction (generally smaller than 20%) of the anomalous rainfall years at any given station show "potential predictability" related to significant non-stationary changes in either the event frequency and/or intensity characteristics for the given year. Investigations of the non-stationary variations in the occurrence and intensity characteristics indicate they display similar significance in explaining the remaining 15% of interannual variance in seasonal precipitation over the region, although numerical tests suggest that the non-stationary interannual variations of these two characterizing metrics are not necessarily independent. Further investigation into the use of these models for identifying "potentially predictable" years related to interannual climate variability will also be discussed.
H51F-05 09:30h
Spatio-Temporal Trends in the North American Monsoon
The North American Monsoon is responsible for as much as 50-70 percent of the annual precipitation in the desert southwest. The variability of this important moisture source has significant municipal, agricultural, and environmental impacts. It is, thus, important to understand the mechanisms driving this variability and to predict it. Results show that there is an increasing trend in summer precipitation over New Mexico and a decreasing trend in summer precipitation over Arizona. This shift has potential links to large scale climate. Correlations with large-scale atmospheric variables indicate that summer precipitation may be more dominated by large-scale patterns set up early in the monsoon season. Relationships with the space-time variability in streamflow of selected basins are also investigated. Links with antecedent winter precipitation and temperature are currently being studied.
H51F-06 09:45h
Climate Variability Impacts on Watershed Nutrient Delivery and Reservoir Production
Reservoirs in agricultural dominated watersheds tend to exhibit pulse-system behavior especially if located in climates dominated by summer convective precipitation inputs. Concentration and bulk mass of nutrient and sediment inputs into reservoir systems vary in terms of timing and magnitude of delivery from watershed sources to reservoirs under these climate conditions. Reservoir management often focuses on long-term average inputs without considering short and long-term impacts of variation in loading. In this study we modeled a watershed-reservoir system to assess how climate variability affects reservoir primary production through shifts in external loading and internal recycling of limiting nutrients. The Bosque watershed encompasses 423,824 ha in central Texas which delivers water to Lake Waco, a 2900 ha reservoir that is the primary water source for the city of Waco and surrounding areas. Utilizing the Soil Water Assessment Tool for the watershed and river simulations and the CE-Qual-2e model for the reservoir, hydrologic and nutrient dynamics were simulated for a 10 year period encompassing two ENSO cycles. The models were calibrated based on point measurement of water quality attributes for a two year time period. Results indicated that watershed delivery of nutrients was affected by the presence and density of small flood-control structure in the watershed. However, considerable nitrogen and phosphorus loadings were derived from soils in the upper watershed which have had long-term waste-application from concentrated animal feeding operations. During El Niño years, nutrient and sediment loads increased by 3 times above non-El Niño years. The simulated response within the reservoir to these nutrient and sediment loads had both direct and indirect. Productivity evaluated from chlorophyll a and algal biomass increased under El Niño conditions, however species composition shifts were found with an increase in cyanobacteria dominance. In non-El Niño years, species composition was more evenly distributed. At the longer time scale, El Niño events with accompanying increase in nutrient loads were followed by years in which productivity declined below levels predicted solely by nutrient ratios. This was due to subtle shifts in organic matter decomposition where productive years are followed by increases in refractory material which sequesters nutrients and reduces internal loading.
http://www3.baylor.edu/~Joseph_D_White/lakewaco.htm