Rutgers Weather and Climate Monitoring and Weather Forecasting System Program Justification

Dr. David A. Robinson and Dr. Richard H. Dunk 31 May 2002

Improving Utility Efficiency and Reliability: A Collaborative Weather and Climate Program between Rutgers University and the NJ Utilities

The Rutgers Weather/Climate Monitoring Network and Weather Forecasting Program is a collaborative effort between The Office of the New Jersey State Climatologist (ONJSC) and the Center for Environmental Prediction (CEP) at Rutgers, The State University of New Jersey. The goal of this program is to provide NJ-area utilities with historical, current and forecast meteorological data in and around New Jersey through an integrated online display system. This document justifies the opportunity for NJ-area utilities to be provided with access to high-resolution meteorological monitoring and forecasting products for improved operational efficiency and security.

The ONJSC operates the New Jersey Weather and Climate Network, or NJWxNet, an hourly, real-time meteorological monitoring network. NJWxNet data is collected, processed, archived, and disseminated for online display via our Strategic Weather Object-Oriented Remote Display (SWOORD) system, providing a one-stop internet resource for New Jersey weather and climate data. (Fig. 1)

Figure 1: Sample of hourly NJWxNet maps under development.

The National Weather Service (NWS) operates ten commissioned Automated Surface Observing System (ASOS) stations in New Jersey that are currently used for statewide monitoring, however the NJWxNet contains approximately 50 stations from several networks with plans to include over 100 hourly stations by 2004. (Fig. 2)

Figure 2: Existing networks to be included within the NJWxNet by 2004. Additional stations in neighboring states are also accessible.

The NJWxNet is fast becoming one of the most dense, hourly, meteorological monitoring networks in the country. This unique collection of hourly data can aid decision-makers in the identification and monitoring of localized meteorological phenomena, such as severe storms, ice storms, and extreme heat and cold. There are at least seven distinct storm tracks known to affect the New Jersey region. (Fig. 3)

Figure 3: U.S. map of at least 7 different sources of storms that affect the NJ region.

These tracks converge over our region, producing the potential for severe storms ranging from snow/ice events to intense northeasters, each with unique characteristics that can adversely impact a utility’s infrastructure.

This dense monitoring network is necessary to cover the diverse climate of urban, rural, coastal and mountainous environments that help to shape New Jersey’s five known climate zones. (Fig. 4)

Figure 4: The five climate zones of New Jersey.

It is impossible for the 10 NWS stations to account for the varying conditions that occur within these local climate areas. The strategically located NJWxNet monitoring sites allow for a more accurate assessment of New Jersey’s climate, which is critical for utilities during extreme weather events (e.g., sustained maximum temperatures above 90 degrees F, prolonged periods of drought, etc.).

Temperature differences in and around the state are often over 20, 30, and sometimes over 40 degrees F at any one time. The accurate location of temperatures between 25 and 45 degrees within a county level is critical for the assessment of rain, ice and snowfall. Severe weather situations, such as ice storms, high winds, or flooding conditions require the activation and mobilization of utility crews and equipment. These adverse situations equate to substantial unnecessary costs if decisions are based on inadequate weather information. The intent of the NJWxNet is to minimize these costs while enhancing the utilities’ reliability and service.

The large number of stations contributing to the network allow for the NJWxNet to observe local conditions at a very high resolution. This high-resolution set of data points will be fed into the Regional Atmospheric Modeling System (RAMS) and modified specifically for utilities in the New Jersey region to produce the Rutgers Predictive Mesoscale Meteorological Model, or RPM3. Models are allowed to be run at a higher resolution when they are initialized with more data points. Studies show that models produce forecasts with improved timing, spatial extent, and severity of events when given a greater number of data points at a higher resolution. RPM3 will be run at a 2km resolution with supplemental input from NJWxNet stations to allow for forecast data to “catch” localized events within individual counties and utility service areas. (Fig. 5)

Figure 5: RAMS horizontal grid configuration used for New Jersey case studies. The model was run using 110 x 170 points at a 2-km resolution for the New Jersey area within the blue box.

Forecast models used by the NWS and private weather companies have a maximum resolution of 12km, which is too coarse to locate phenomena such as the coastal sea breeze, an isolated severe storm, or the rain/snow line. Case studies are provided that compare RAMS forecasts with forecasts provided by the NWS “ETA” model and actual observations. It should be noted that although the RAMS forecasts show a substantial improvement over the NWS forecasts, further improvement will be realized when RAMS is refined to become RPM3. The configuration and capabilities of RPM3 will prove to be superior to any other currently available prediction model. RPM3 will be a unique forecasting tool for utility operations.

Along with near-surface data, RAMS utilizes “upper-air” data to account for the three-dimensional dynamics of the atmosphere. The NWS upper-air network does not include a station in New Jersey. However, upper-air data specific to the New Jersey region will be incorporated into the NJWxNet from two locations. These sites were funded by PSEG and GPU and are owned and operated by Rutgers. They are strategically located in central NJ and along the NJ coast in East Brunswick and Tuckerton, respectively. Access to this additional upper-air data in New Jersey will serve to enhance the ability of RPM3 to accurately simulate local conditions.

The case study documents highlight both severe and extreme weather events chosen by utilities representatives for their adverse affect on NJ utility services. These events include the 4-6 March 2001 “Storm That Wasn’t”, the 9-10 August 2001 heat wave and severe storms, and the 19 April 2002 severe storms.

The RAMS model output is not archived for case study purposes at this time, however we have obtained precipitation output for the 4-6 March 2001 event and temperature, wind and precipitation output for the 19 April 2002 event. The ETA model is used for comparison as the standard forecast model used by the National Weather Service and private meteorological consultants.

These case studies present strong evidence of the individual strengths of RAMS and the NJWxNet. Forecast improvements are expected with further refinements to RAMS, which will include the integration of NJWxNet data to create RPM3.

4