Click for next page ( 20


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 19
3 Modernization of the National Weather Service Hydrologic Services: An Evaluation Four major aspects of the production of hydrologic and hydrometeorological forecast products within the modern- ized NWS provide a means to evaluate each component of the operational hydrology program. Four main sections in this chapter correspond to these aspects of forecast produc- tion. First, the forecaster uses observation inputs to produce forecasts and warnings; the availability and accuracy of these observations determine, to a large extent, the quality of the hydrologic forecast. Second, the forecaster relies on a set of forecasting tools and techniques to develop the required products. The manner in which data are collected and inte- grated into operations is analyzed in a third major section. Finally, the products and services delivered to the user com- munity and to the general public are evaluated. OBSERVATION IN PUTS One of the most central aspects of hydrometeorological operations is analyzing, estimating, measuring, and forecast- ing precipitation in the form of both rain and snow. Observa- tional data come from a variety of sources, including NEXRAD, satellites and aircraft, and surface gauges (includ- ing the ASOS [Automated Surface Observing System]) that are read and reported by a network of observers and auto- mated devices. Precipitation Processing System The goal of precipitation data processing at the NEXRAD site and at RFCs (River Forecast Centers) is to define the concise spatial distribution of precipitation over appropriate time intervals. This spatial and temporal distribution of pre- cipitation is designed to serve as an input to continuous simu- lation hydrologic models. The NWS uses both rain gauge data and radar estimates of precipitation as inputs to a three- stage precipitation processing system, or PPS (see Box 3-1~. Other information available to the forecaster, such as satel- lite imagery showing the are al extent and intensity of pre- cipitation, can be incorporated in Stage III of the PPS. 19 Scientists have been using weather radars for nearly five decades to estimate the amount of precipitation that falls to the Earth's surface. Although radar technology and the asso- ciated computer processing have improved immensely in recent years, our knowledge about the amount of precipita- tion that actually reaches the surface remains primitive. There are numerous reasons for this shortcoming, including the outdated scientific basis of the design of the PPS, funda- mental limitations in the performance of the NEXRAD, implementation flaws in the parameterization of the PPS, and operational restrictions from inadequate data transmis- sion or computational facilities. The net impact is that most radar specialists recognize that the existing NEXRAD rain- fall algorithms are outdated and seriously flawed (e.g., Atlas et al., 1996~. For example, the current PPS design has essen- tially no features to detect or remove the effects of so-called bright band contamination due to the enhanced reflectivity of radar returns from melting snow. As another example, the current PPS implementation uses a single Z/R relationship (see Box 3-1) for most of the country regardless of the geographical region, season of the year, or time of day- which is counter to the very science that developed the de- fault relationship and results in its use in situations that were not represented in the original developmental data. As a third example, the Stage I PPS design incorporates a procedure to include up to 50 rain gauge reports at each radar site so as to calculate a single multiplicative bias, updated hourly, which inflates or deflates the final precipitation totals in an attempt to improve the accuracy of the PPS. Systems to actually deliver these rain gauge reports to operational systems have not been implemented. As a final example, the PPS includes a Stage II processing procedure that performs a sophisticated merger of rain gauge and radar data, in effect a localized bias rather than a single value for the entire radar. But Stage II processing is not yet available at WFOs (Weather Forecast Offices) because of delayed AWIPS (Advanced Weather Interactive Processing System) implementation. Considering these limitations, it is not sur- prising that experience in NWS field offices shows that the

OCR for page 19
20 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ....................................................................................................................... ............................................................................................................................. .. ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::: :::::: :.:.:.:.:.:.:.:.:.:.:.:.:.:.:-:-:-:-:-:-:-:- :-:-V:V:~:-J.: :J-l:-:l-1-:~:-1f.~:~:-:V:. A: :f~:~: At:: ~B:J:~:~:I:': ~ A:: rs l: :~:: :. :J:: :l :l: l::: t :l: lo:: A: I: I l: :l: ~-l: ::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::-::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:.-: - :-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-:-:'-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:1-:-:-:-:-:-:-:-:.-:-1:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: Ilme !nlewa!s as soon lney are represenlallve o-n--ly~ ~---ln-e~ g-a-u-g-e--s~ !-m-m-eo--l-ale~ e-n-,.v,l-mn-m-enl~ Kq-l-n~ g-a-u-g-e~ n-e~o-rKs :::::: . :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:'.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.~:.:.:-:-:-:-:-:-:-:-:-:-:- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::I::::::::::::::.:::l::::::::::::::::::::::::::::::::::::::::::l:::::::::::::::::::::::::::::::::::::::::.:: :::::::::::::::::::::::::::::::::::::::::::::::::::::::J :::::::::: p-rovl-a-e~ ln-ese~ o-ose- a-l-l-o-n-s~ are own-ea~ Ll~ en m-a-n-ag-ea~ an-a m-a-l-nla-l--n-e-a)~ ~..~ a n-u--m-De-r~ or coop-e- a-l-l ::::: :-:-:-:-:-:-:-:-:-:-:-:-:- :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-1:-:-:-:-:-:-:-:-:-:-:-:-:1 -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-:-:-:-:-:-:-:-: Tl-:-:-:-:-:-:-:-:-:~-:-:l-l -:-~-:-/:~-:-:-1-:-:-:-:-:-:-:-:-:-:-:-:1 - - -:-:-:-:-:-:-: - : -: . -----a-n-a~ l-oca --l~ ag-en-cl-es~ l---n-e~ l~-vv~ n-as~ l-!--m-l-leo~ ~ :::::: :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:,,, :.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: :::::::::::::::: .:: ::: ::::: _1 ::::::::: .::: ::::::::::::: ::::: ::::: :- - - - - - - - - - - - -A -I. -q-~-l -- - - -a- ~ Hi- - - it- =-l --it ~-= ---- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ::::::::::::::::::V:~1:~:~I:V I::!:::U:1:1:~:::!:1::!~1:1::!:~1::!~1::!~ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:1-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: - - - - - - - - - - - - - - - - - - - - - - - -A - - - - - - - 1- - - - - - - - - _- - - -, - .- - - - - - - - - - - - - - - - - - - - - - - - -I- -am- - - - - - - - - - .- - - - -.- - - - - - - -. - - - - - - - - - - - -I - -l -_ -- ~ Kaa a r--ts-uma- ~s~ or--- ~p ~ra-uon~ !~--tA-~-u~ p~l-a-es~ a-ma-l es~l--m-ams~ old p~cl-p-!-la -l-l-o-n~ o-wr---a-ls-crele~ --m-e lunge- a-!-s ................................................................................................. ....................... ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::.:::-:::::::::::::::::::::::.:::::::::::::::::::::::::::::::::::::::::::::::1:::::::::::::::::::::::::::::::::1::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::~:::::: - : esll--m-aws~ are oas-ea~ on an e-m-p-l--rl-c-a-l~ co-n-ve-~-l-o-n~ l--n-lo~ ra-l--nral-l~ a-les~ or ln-e~ l ..... ....... ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-' -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:1-:-:-:-:.-:-:-:-:-:-:-:-:-:.-: 1 :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-:-:-:-:-:-:-:-:-~-:-:-:-:-:-:-:-:-:-:1-:-:-:-: ~-een~ relu--rn-ea~ lo ln-e~ ra..-a-a-r~ tours ~-ac~-cane-re-a)~ T-mm~ p-rec-l--~-llall-on~ an-a oln-e-r~ pa-rl-l-c-l-es~ A s-l--n-g-l-e~ conversl ............................................................................................................ . ............................................................................................................................. :::::::::::::::::::::::::::::::::::::::::::::::::1::::::::::,:::::::::::::::::::::::::::1::::::::::::::::::::::::1:::: :l::::::::::: :1::::: :::::::::::::::::::: 1:::::::::::::::::.:::-:::::::::-::.:::::::::::::::: ::: :::.:: re-l-ales~ c-l-o-u-a~ ~--ro-~-l-el~ rerl-ecll-vl-~ ~ lo s-u-~ac-e~ a-l--n-ra-l-l~ ~ l-n-m-u-g--n~ an em--p-l-rl-ca-l~ wl-all-o-n-s-n-l-p~ Known as---ln-e~ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::: ::::: ::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-:~-:-:-:-:-:-:-:-/~:-:-:-:~:-:-:-:-:-:-:-:-~:-:1-:-:-:-:-:-:-:-:-:-:~-:-:1-:_-~-~:-~-:-~:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:':-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- - -:-:-:-:-:-:-:-:-:-1:-:-:-:-: ........................ snip. ln Toe its - u Item a single ~ :::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::~::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: Rae -tA~ cove age areas Holmes In l~plcal cllmams use a alTrerenl walk value.) ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::T:I:::::::::::::::::::::::::::::::::.:::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::.:::::.:,::::: :.::::::::::::::::::::: I ~ r-~^ Or ~ ~ ~ ~ Fir ~ reel ~ l r Ad Tl:~:~::::::~:r~:~:~:~.:~:~:~::::::~:~:~:rl:~::::T:~::~:::::~.-:l:::l::rr^:~:T:::::~.:~:`rT:~:~: ::::::::::::::::::::::::::::::: :':':':':':':':':':':':':':':':':':':':':':':':':':':':':':!':':I':'I:I:'=~':':'a:~: - '~':':'V:!':':':'~:!':'~!': - 'l:'~"'~!':~.:'l:':!':':':~'l:':V`~aat'l:':!':~:':':'~V'I':l:'I'~:!':'l:~=':':'~!':'l:=':':'=~:'l:':!':'~'l:':!':~:':'a:Y: O'~:!':'l:' ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:~-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-1:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:1-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:1-:-:-: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::: ::::::::::::::::: ::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::: 'St' ' ~ I'-'-'p~ ,,t-'-'t'b'' ~ ' ' d'' '''-'-' 'at' ~d'-'-'-'' ' '' ' '-'-'N'EX~D'-'-'' ' " '' " -t' ~ '' t' -' ' '''-'-'T'h'i' '-'-'' t' ~A--- j .................................................................................................................................... ............................. --- - t~ I-i-- -'it- -do to fax i d-- to to d- i fi-- -I-d by fax t-h ~do tin ~ A ....................................................................................................................... ........................................ - - - - - . co.n.tro.l~ ~-- .............................................. ................................................................................................................ .............................................................................................................................. --- ---- - - --- St~ 1~ i--1- be ~d~ tithe W- -- the Fax t~Offi ~Adz Amp- the 1~ ~Fit P- ....................................................................................................................... ::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::: 'S' ' t' ' ~ p'' 't' '' -'-'-' '' '' ' '-'-t-b ~' al'l-' d'''-'-'Add-'i-ti'' ~ 1~' ' 'I 'd-' t' 'I '' '' ' do k' ~ 1'-i'W'~ ' t' ' -1'~b Kit the -' Id ................................................................................................................................. -- - - ::::::::::::.:.:.:.:.:.:.:.:.:.:.:.:~:~.I:.:I.r.-~:~:.:.:~:q.:l.:.l:~'' '' ':':':q'' ':~:':':'r.'' :~':q:'r-':':' :' ~ '' :q'l'' '--~--q Rae 'I.-' :' ~':':':~':'i':':'q:'' ':':':q a-~-u-l--~ \y-~-u-y-~J~ us----- l--~-~-U-I~ ~a-tl--l---l---l-~-~aJ~ -A ~ l ~ lull you By al l Am l mu, u-l---l-u-l-y-a-l-~ ----v ....................................................................................................................... ............................. -. -. - -. -. -. - - - . '-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'-'St' ' ~ Ill'-'-'p-' ' ' ' -i"' " """' ' ' ' " '-'-'' 't'-'-'R-F'~' '-'-'' h .................................................................................................................................... - ---- --- ---- -- - - - - tb~ R-FC~ k " ~ b-i-~-i- ~1-- d~ i-- t ~i 1 ~j ~ JO k --oh the - ~ ....................................................................................................................... .......................................... -. - - -. - - - ' t' ~ it i tuba ' 1~ '' t' 9'' '-'-th-' t-'-'i ~At' -' ti' -'-'' '' 'Dot 'the fib it ;~ tuba ' 1' '-'' a-9'' 'I ' h';' he tuba he k ..................................................................................................................................... - -- - -- .................................... h a -.-s - . . .. .............................. - - -. - - - - .......................................... .................... .......... .................................... .......................... ............... ...................................................................................................................................... ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: PPS provides inconsistent results for certain types of rainfall in different geographical areas and has seasonal variations as well. Conclusion. Successful modernization of NWS hydrologic services depends on the ability of the NEXRAD network to provide accurate estimates of precipitation that benefit from improved spatial and temporal resolution. Furthermore, the current approach for analyzing precipitation patterns (which includes the computation of a single bias for each radar and the use of a single ZIR relationship for most of the country) is primitive and flawed. It lacks the scientifically sound and dynamic methodology needed to improve the seasonal per- formance of the PPS in light rain or snow and in mountain- ous areas. The Committee on Meteorological Analysis, Prediction, and Research (NRC, 1994) recommended that "the NWS [should] work to establish the accuracy of NEXRAD rainfall estimates, especially for heavy rainfall events." The present report reemphasizes this recommendation as an area needing additional scientific investigation and study. The committee further recommends: Recommendation 3-1. The NWS should continue its efforts to incorporate additional real-time precipitation data into hydrologic products and services. The methods used for multisensor detection and estimation of precipitation should enable accurate characterization of precipitation patterns that span seasonal, geographic, and range diversity. A capability to distinguish reliably between rain and snow must be devel- oped. The PPS methodology should be upgraded to a more scientifically sound and dynamic methodology to improve seasonal and geographic performance, especially during light rain and snow events and in mountainous areas. Finally, important details in the precipitation patterns from NEXRAD are not currently available to meet certain needs of the flash flood program, especially for flood-prone basins that are as small as 1 0 square kilometers. For example,

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES NEXRAD precipitation initially is acquired in a polar coordi- nate form (azimuth and range in 1-degree x 1-km increments) with 255 data-level resolution. But NEXRAD products are "coarsened" either in spatial detail (4 km x 4 km x 255 data levels for the hourly digital product) or in both spatial detail and data resolution (2 km x 2 km x 15 data levels for other graphical products). The details lost can be critical to the timely issuance of flood warnings in rugged terrain. NEXRAD rain- fall accumulation, determined at the highest resolution of the radar, represents an information data source that would permit the issuance of timely flood warnings in very small basins. The committee understands that, in software Build-9, the NWS plans to produce all graphical NEXRAD precipita- tion products in a polar coordinate format and to add a digi- tal hybrid reflectivity product with high resolution. Build-9 is expected to be released in the fall of 1996. Conclusion. In coarsening radar precipitation data for op- erational use, the NWS has been missing an opportunity to exploit the full power of its new NEXRAD technology for use in the WFO (Weather Forecast Office) hydrology pro- gram. The committee applauds the plan to shift to the use of polar coordinate format for NEXRAD precipitation prod- ucts. This change represents an important improvement in precipitation processing for flash flood forecasting. Recommendation 3-2. Rainfall accumulation maps that use the highest resolution possible from NEXRAD should be made available to support the flash flood program. Although there are important areas that need improvement, the committee is encouraged by the progress made by the NWS toward improving PPS performance, as represented by new software builds for NEXRAD, reprogramming of Geostationary Operational Environmental Satellite data col- lection platforms, and attempts to link the NWS to new sources of data. Quantitative Precipitation Forecasts NEXRAD provides a valuable opportunity to improve flood forecasting and management of water resources. De- tailed QPFs (quantitative precipitation forecasts) represent another opportunity to significantly improve both flash flood prediction and regional runoff estimates that, in turn, impact hydrologic forecasts for larger basins (Davis and Drzal, 1991; NRC, 1991a). However, the production of an accurate QPF is consid- ered to be among the most difficult challenges in operational meteorology, even when sophisticated computer models of the atmosphere are used to produce guidance information for the human forecaster. The challenge increases substan- tially when hydrologists attempt to use the QPF as part of the precipitation input to their hydrologic models, which re- quires forecasts of precipitation spatial coverage and amount. As a result, the accuracy of current QPFs is largely 21 a function of the skill of the individual forecaster and varies considerably from one forecaster to another. The complexity of this hydrometeorological forecast problem begins with the physical space and time scales on which precipitation occurs. For example, it is a relatively easy matter to develop a moderately accurate, 24-hour fore- cast of precipitation when the verifying amounts will be less than an inch or so. But when the atmosphere organizes itself to produce an extreme, often destructive precipitation event (more than 10 inches in less than 6 hours), meteorologists have almost no skill even 6 to 12 hours in advance to correctly anticipate these highly focused events. The fore- casting difficulty derives from the fact that the physical pro- cesses that occur in the atmosphere during these extreme events are not well understood. Moreover, these processes occur on small space and time scales and thus escape detec- tion by the current federal observing networks. One purpose of an accurate QPF (in both space and time) is to extend the warning lead time of a critical hydrologic event. But the range of uncertainty in space, time, and pre- cipitation volume in a modern-day QPF can create a projec- tion of runoff in a specific basin that ranges from minimal to excessive. Even as hydrologists seek to accurately partition past precipitation into the critical component of actual run- off in a specific basin (with its variable terrain, soil type, and land use), the hydrologic forecaster must also answer an- other set of difficult questions; namely, in which specific basin will the forecast precipitation fall, at what rate, and over what preexisting basin conditions? In other words, the impact of uncertainties in meteorological observations as well as with complex scientific problems in the meteorologi- cal forecast process is amplified into larger uncertainties in the resultant hydrologic processes. These severe limitations, coupled with an almost nonexistent ability to accurately fore- cast specific amounts of precipitation in a specific basin, define the difficult task of accurately forecasting water lev- els in the streams and rivers of a basin. It is this situation that justified the establishment of hydrometeorological analysis and support units at each RFC. Conclusion. Although the accuracy of QPFs has improved over the past decade (Olson et al., 1995), this improvement has had a minimal impact on the improvement of hydrologic services. Thus this current report reemphasizes a previous NRC recommendation (1991a), namely: Recommendation 3-3. Incorporation of improved QPFs and associated uncertainties into the hydrologic models for short- range and long-term stream-flow forecast is essential and requires collaborative scientific investigation by the NWS and the academic community. Both QPF and probabilistic QPF (pQPF) are perceived by many in the NWS hydrology program to be of great value for the development of improved and long lead-time hydro- logic services (see, for example, Krzysztofowicz, 1995~.

OCR for page 19
22 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES Moreover, the pQPF methodology represents a new thrust designed to enhance the utility of the resulting hydrologic forecasts by placing more timely information into the hands of users of hydrologic data. Nevertheless, substantial im- provements are needed. Conclusion. Significant efforts are required to coordinate the possible redesign, production, and use of QPFs. In addi- tion, the impact of a modernized QPF on hydrologic models in various geographic and seasonal conditions needs to be assessed. Both sets of forecasts need to undergo extensive verification studies to determine their proper design for a modern-day QPF (e.g., required time projections and spatial resolution for hydrologic models of the twenty-first century before QPF input becomes routine). Recommendation 3-4. The NWS should accelerate its fledgling efforts to redesign, develop, evaluate, and verify QPFs and pQPFs and assess their use in hydrologic forecast models across a range of geographic and seasonal condi- tions. The Office of Hydrology should determine the time and space resolution of QPFs that hydrologic models require. Users of products who incorporate QPF data should be kept informed about these developments and their potential im- pact on user operations. Snow The ability to measure snow depth or to model snowmelt (especially in remote mountainous areas) represents a challenging hydrologic problem. The current methods are inadequate. In the first place, it is difficult to determine snow depth and water equivalent of snow over large areas to a reasonable degree of accuracy. In fact, snow depth or its water equivalent is much more poorly estimated by NEXRAD precipitation data than is rainfall. Progress has been made in recent years by the NWS National Operational Hydrologic Remote Sensing Center (NOHRSC), which has used aircraft and satellite data collection methods to improve measurements of the areal extent and depth of snowfall (Carroll and Holroyd, 1990~. Conclusion. Through its efforts, the NOHRSC has advanced the capabilities to produce guidance on areal snow coverage and snow-water equivalent. Critical measurements provided by the NOHRSC over large areas during the winter season cannot be obtained as effectively by any other technology. Recommendation 3-5. Adequate resources should be pro- vided to continue areal snow surveillance and to maintain the full complement of remote sensing technologies and ac- tivities now provided by the NOHRSC. Information from these NOHRSC activities and the use of their guidance prod- ucts should be fully integrated into RFC operations. Current snowmelt techniques are derived empirically and are not based on physical principles; the techniques are based on degree-day computation, a cumulative index related to a specific temperature threshold. As a result, during conditions of rapid melting, most hydrologic forecasts underestimate the volume of water that reaches streams and rivers. Thus flood forecasts and warnings tend to be inaccurate during periods of rapid snowmelt. The volume of water and runoff created by melting snow often makes a critical contribution to flood conditions. Conclusion. Improvements are needed in physically based scientific techniques to predict snowmelt while retaining the positive aspects of empirically based hydrologic methods. Recommendation 3-6. Additional resources must be de- voted to improve the scientific bases to monitor and predict snowmelt, especially during situations that involve rapid melting. Surface-Observing Networks Surface-water observations are made by stream gauges that measure water level in rivers, streams, lakes, and reser- voirs and by precipitation gauges that measure rainfall and the water equivalent of frozen precipitation (snow, hail, etc.. Both stream and rain gauge data are critical sources of input data for NWS hydrologic models. Unfortunately, evidence suggests that the surface-observing networks that support hydrologic operations in the NWS and in other federal agencies are deteriorating with each passing year. There are different issues associated with each type of data network. The backbone of the national stream gauge network is an intergovernmental network of continuous and partial-record gauging stations on the nation's streams, lakes, and rivers known collectively as the Cooperative Stream Gauging Net- work. The U.S. Geological Survey (USGS), which has a ba- sic mission to collect surface-water information in the United States, administers the cooperative network and operates more than 85 percent of the stations in the network that sup- port hydrologic operations in the NWS and in other federal agencies. The network is supported by funding from numer- ous federal, state, and local agencies. The USGS provides cooperative funding (up to 50 percent of the total) for ap- proximately 63 percent of the national stream-gauging net- work; the USGS is the sole funding agency for approxi- mately 6 percent of the national network. Water-level data from these stream gauge stations are a critical input to the NWS hydrology program. For RFCs, the stream gauge locations determine service locations and model control point locations. Data from these reporting stream gauges are used for river forecast generation, updates, and verification. Because of reductions in funding support, the number of stream gauge locations in the national stream-gauging network Precipitation gauges are generally referred to as rain gauges.

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES has become gradually smaller over time. During 1983-1994, 86 NWS service locations were affected by closures at USGS stream gauge locations; this represents 2 percent of all NWS locations nationwide. The rate of closures has increased rap- idly since 1989. If model control points are included (i.e., USGS stream gauge locations used in RFC hydrologic pro- cedures), the number of locations affected increases to 193. The decline in financial support has accelerated over the past five years because of tightening federal, state, and local budgets. This trend is disturbing because agency budgets likely will continue to tighten as pressures mount to reduce budget deficits. Conclusion. Because the NWS does not financially support the operation of the USGS stream gauge network (except for staff gauge locations in the NWS cooperative observer net- work), the NWS has had a minimal influence on the place- ment, operation, maintenance, or potential termination of USGS stream gauges. Issues associated with the rain gauge network are related not to any decrease in the number of rain gauge locations, but to the quality of data from cooperative networks, the availability of rain gauge data in real time, the geographic distribution of rain gauges, and the temporary loss of accu- rate precipitation data from locations where the ASOS is used to obtain local climatological data. In turn, the latter issue impacts the continuity and accuracy of the historical precipitation record. Observations from rain gauges are available to the NWS from a variety of reporting networks. The location of these gauges is governed by the requirements of the owner agency. Many of the rain gauges are owned and operated by NWS partner agencies that share data with the NWS. These pre- cipitation observations are especially important in the con- tinuous modeling of soil moisture, snow accumulation, and snow ablation and in the computation of runoff in NWS hy- drologic models. The largest reporting element in the rain gauge network is the NWS cooperative observer network. The current coop- erative observer network operates in basically the same man- ner as it did at its inception over 100 years ago. Although the network has been very successful at fulfilling its original mission in agriculture by defining the weather and climate of the United States, its data are now being used in a wider variety of ways, including precipitation processing with NEXRAD and longer-term forecasting of water resources. The majority of rain gauge readings from the cooperative network still are acquired manually and thus are updated at most only a few times a day, based on NWS reporting crite- ria. The observations produced manually from the coopera- tive network are transmitted either daily or monthly to NWS offices by telephone or in writing on a standard form. A primary deficiency of the cooperative network is its antiquated technology, which reduces data availability or results in inferior and missing data. In addition, the network 23 is becoming less stable because of the growing rate at which new volunteer observers must be recruited to replace ob- servers who are retiring or relocating. Finally, the observa- tions are acquired at different or changing times across the United States, which makes for a very cumbersome process for using the data in the modernized NWS. Of the 10,600 volunteer weather observer stations in the cooperative network, about 9,500 support hydrology require- ments and make up the "hydrologic" portion of the network. About 5,000 of the 10,600 volunteers support the "climate" program.2 Approximately 1,000 of the hydrologic locations provide real-time gauge readings (either rain gauge or stream gauge or bosh) to support modernized hydrologic operations; these can be interrogated either by phone or by satellite. The majority of the cooperative observer locations, however, can not provide real-time rain and stream gauge readings because their gauges cannot be interrogated by either phone or satellite. Another large subset of rain gauge observations is ac- quired on a four- to six-hour basis from data collection plat- forms by way of satellite interrogation. The main concerns regarding these gauge observations are the need to receive the data in real time and the uncertain quality of the precipi- tation observations, especially during frozen precipitation events. The NWS is currently working with the primary owners of these data collection platforms to acquire real-time gauge readings by reprogramming data collection platforms for random transmission. Some community (e.g., local flood warning systems) and state networks provide their rain gauge data to the NWS on a real-time basis; but here, too, the quality of the precipitation observations is sometimes questionable, especially during frozen precipitation events and high rainfall intensities. Data quality is predominantly affected by the type of rain gauge and the frequency of maintenance (i.e., routine inspection and calibration) performed by the owner agency. Clearly, data quality from these local networks is beyond the control of the NWS. The majority of rain gauges used by state- or locally- owned networks are "tipping bucket" gauges. This type of rain gauge is often deficient during high rainfall intensities and frozen precipitation. Some local networks have added rain-rate correction equations to their data processing sys- tems to account for errors during periods of high rainfall rates; however, few owner agencies have actually compared the corrected observations against a reference-standard rain gauge. In addition, each owner agency has its own preven- tive maintenance (i.e., calibration or adjustment) schedule for its rain gauges; some owner agencies do not perform any calibration on their tipping buckets. This situation obviously will affect the quality of and limit the use of the precipitation observations in modernized hydrologic forecast operations. NEXRAD precipitation estimates are among the main sources of data for hydrologic forecasts. Remotely sensed 2Some cooperative stations support both programs.

OCR for page 19
24 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES rain gauges contribute significantly to the NEXRAD pre- cipitation processing system by providing adjusted radar pre- cipitation estimates to NWS forecasters. The NWS hydrol- ogy program has performed some initial assessments of the rain gauge data support for NEXRAD precipitation process- ing (Seo et al., 1996~. In general they found that there is a strong need for additional rain gauges in many parts of the United States, including the upper Midwest, the West, the Southeast, the Southwest, and the Northern Plains. These additional rain gauges could be installed at locations in ex- isting networks that presently do not report precipitation data, especially on data collection platforms owned by the NWS and other agencies (provided the site exposure is ad- equate for precipitation measurement). There is also a need to acquire additional real-time re- porting rain gauge data throughout the nation. Given the spa- tial distribution of the cooperative observer network nation- wide, upgrading all nonrecording rain gauges in the network with rain gauges that can be automated and equipping all existing automated weighing rain gauges in the network with satellite, phone, or radio interrogation capabilities would greatly facilitate NEXRAD precipitation processing. A continual effort must be exerted by the NWS to ensure the flow of rain gauge reports for use both in the calibration of precipitation estimates from NEXRAD and in the genera- tion of river forecasts from hydrologic models. The simple observations represented by stream gauge or rain gauge read- ings are perhaps the most critical input data to hydrologic models. Conclusion. The modernization of NWS hydrology will be hindered without the real-time availability of stream and rain gauge data. It is important for the NWS to take firm, aggres- sive steps to continue strengthening this major weak link in its flash flood and river flood programs. The following are three vital areas that need attention: (1) improve the NWS' ability to efficiently retrieve observa- tions from the cooperative observing network, from gauges owned by the U.S. Army Corps of Engineers and the USGS, and from satellite-interrogated data collection platforms; (2) work with other federal, state, and local agencies to im- prove the geographic distribution of real-time reporting rain gauges in data-sparse regions, so as to provide adequate rain gauge data support for each NEXRAD coverage area; and (3) strengthen coordination activities with other federal, state, and local partners to improve the quality of precipita- tion observations and other shared information. Recommendation 3-7. The National Oceanic and Atmo- spheric Administration (NOAA) should review the status of the cooperative observing network and plan for its future in the context of the ongoing modernization.3 3This committee is currently coordinating with the NOAA to initiate a study of issues related to modernization of the cooperative observing network. Recommendation 3-8. Additional rain gauges should be put in place in many areas of the United States where hydro- logic data are sparse. These gauges should be installed at locations in existing networks that presently do not report precipitation data. To the extent possible, these should be automated, real-time reporting rain gauges equipped for re- mote interrogation. Recommendation 3-9. The NWS should strengthen its part- nerships with other agencies by contributing to the financial support of telemetered networks that are critical to public safety. Inaccuracies in precipitation measurements from the ASOS rain gauge have been documented by the Office of Hydrology and by the NOAA Climate Data Continuity Project conducted by Colorado State University (McKee et al., 1996~. Since 1991 these inaccuracies have affected the continuity of the historical precipitation record (especially during frozen precipitation events) at locations where ASOS records and reports local climatological data. The Office of Hydrology arranged to install and monitor a universal rain gauge at all WFOs collocated with an ASOS site so as to maintain the accuracy and continuity of the historical pre- cipitation record for a small subset of ASOS locations that also report local climatological data (i.e., approximately 48 locations). However, the remaining ASOS local climatologi- cal data locations will not provide the same data quality and accuracy during frozen precipitation events. Engineering modifications to the existing ASOS instruments are expected to provide more accurate measurements of liquid precipita- tion at all ASOS locations. It also is expected that accurate inputs to the historical precipitation record will resume at all local climatological data locations after a new all-weather precipitation gauge is installed. TOOLS AND TECHNIQUES National Weather Service River Forecast System The NWS River Forecast System (NWSRFS) represents a major operational and technological advance for the mod- ernized NWS hydrology program. It is an interactive data management and forecasting software environment that re- places outdated tools and techniques used in RFCs (Page and Smith, 1993~. See Box 3-2 for a technical description. In developing the NWSRFS, the NWS Office of Hydrol- ogy has led the effort to bring advanced computation and communication technology to hydrologic forecasting. The development of the NWSRFS also represents an important initiative by a NWS team determined to move ahead with modernization plans despite delays in development of the AWIPS. As a result, the hydrology and hydrometeorology components of the AWIPS software system are among the first to be tested in the NWS. The availability of the NWSRFS at all RFCs has generated

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES 25 H- . , sync trot O91C ~-ro-ce-s-s-e-s~ ~--o-c -e- -e-c ~ ~ lye t- ~-e~ ~ ~ --at-l-o--n-a- ~ w-eat- -err ~-e-~-l-ce~ t~ errs --o-- e-cast ~yste-m ... ..... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: T1e ~ -a -~-o-n-a- ~ vv-e-a ; her ~ewIe -v-e- --o-rec-as ; lays semi ( ~-VY-~'Kh-~J'~ I'S ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. h d 1 i ~ 1 ~ i fl - - di i - - ...................... .................... ........ ........................................................................... -:-:-:-:-:-:-:-:-:-:-:-:---:-:-:-:---:-:-:---~-:-----:-1:---:-:1-:-:---:-:---:-:-:-:-~ :-:---:-:-:---~-:-:---:-:-:---:-~:-:-:---:-:-----:-1:-1:---:---:-:-:-:-:-:-:---:-~-:-:---:-:~-:---:-:1-:1-:-:-:-:---:-:-:---:-:-:---:1-:-:-:-:---:-:-:- va-~-a-~-l-es. ip-rl--~-c-l--pa-l--ly.. r.-a-l-nra-l--l. an-a. s-n-om-a-l-~. n-ea-~su- ace. a-l-r. le-m-p-e-~.-alu-~. n-u-m-l-a-l-~. an-a. * .-l-n-a. s-p-eea-~. j i i h h i ~ ~ ~ h i- - hi - di i ~ d - di h d ................. ........................................................................................................... ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:---:-:-:---:-1:-:---:-:-:---:-:. -:-:---:-:-:---:1-:-:-:-:-:---:-:-:---:-:~ :-:---:-:-~:-:-:---:-:-:-:-:---:-:-:-:---:-:-:-:-~:-:1---:-:-:---:-:-:-:-:-:---~:---:-:-:-:---:-:-:---:-:-:-:-:---:-:-:---:-~-:---:-:-:-: s-e-l-e-~e-~ pO-l-n-ls~ o-n~ ln-e~ rl-v.-e-~ n-el-w.or~ ~ ~e ~ a-m-~-n-g~ ln-e~ p-rocesses mooeleo Dy lne I~VV~Kh~ IS lne :::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: i ~ di i ........................................................................................................ ..................... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ''''''-:-:-:-:-:-:-:-:-:-:-:-:-:~-:-:-'-:-'-'-:-:-:-'-: I :-'-:-:-'-:-:-:-:-'-:-'-:-:-~:-:-:-'-:-:-:-:-:-'-: I :-'-:-:-:-'-:-:-'-'I-:-:-:-'-'-:-:-:-'-:-:-:-'-I:-:-:-:-:-:-:-:-:-:-:-'-:~-:-'-:-:-:-'-:-:-'-'-:-:-:-:-:-:d-:-:-'-'-:-~:-:-:-'-:-:-'-:-:-: Iracx* -a-1e~n-ea~a-n-~.s-u-~ace---slaled Inalconlrol a-1-n-ra-1-1- runo~lransTormallon Inel~vv~Kh~al$oinluced snowpac ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 1 d I'i"' '' """' ' ""' 1'1"""' ' """ i' ' ' '' ' ' i"' " """' ' " " i"' " ' """"1" """i" ' ' " ' ' ' i" ' """ ' ' ' ""'d i" "' """' 1'1'' " ---f -' ' ' '' ' -' ' -'-'' ' -'-'-' ~' ' i' ' '1' "' ................................................................................ ............................................ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: in-e~ m-o-a-e-l~ l--~ led 1lng lne lmpacl oT allorallon ana unce~aln parameler$ ano lnpuld ~ a-s~ we-l-l~ as~ l :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::: :::::::::::::::: ............................................................................................................................. '"""""""""""""""""' ' ' ' ' ' """'' ' """'' b '' ' ............................................................................................................................. :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:-:---:-:-:--- - -:-:---:-:-:---:- - -:-:---:-:-:---:-~-1:---:-:-:-:-:-:-:---:-:-:-:---:-:-:-:-:-:-----:-:---:-:-:-----:---:-:-:-----:---:-:-:-:-:-:-:---:-:-~:-:-----:-----:-:~-:~ :-:---:-:-:-:---:-:-:-:-:*:-:-:---:-: p-~-n-cl--pa-~-ly. o-n. com-m-u-n--l-call-o-n. le-c-n-n-ol-og--l-e$. a-n-a. l-n-e. l-nle-~.-acil-Ve. TomcasilOg. c-a-p-a-~-l-l--l-l-l-es. l~-o-n-eln-el-ed :::::::::::::::::: :::::::::::::::::: :::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: '"""""""""""""""""'' "I'''i" '' ' ' '1'' """ I'i" """' " """" '' ''' ' ' ' i' ' ' i'' " " """' f""'h' d"' ' '1'' ' "i" """' " ' ' ' ' ' ' ' """' '' 'd"""' h"' """ '' '' ' ' ' ' ""'i" """' ' '' ' ' "' i"' '' """' h'' """' 'h' 'i'' ' ""'~""' -:-:-:-:-:-:-:-:-:-:-:-:-:-:---:-:1-:---:---:-:-:-:---:~ :-:---:-:-----:-1:-:-:-:-:-:---:-:-:---:-:---:-:-:-----:-:---:-:---:-:---:-:-----:-:---:-:-:-:-:---:-:-:-:-:-:-:---:-1:---:-:-:-:-:-:---:-:-:---:-:-:-:-:-:-:-:---:-:~-:-:---:-:-~:-- p-nysl-ca-l~ prO-ces$es~ d u-cn~ a-s~ ra-l--n a-l-l-- ru-n-on~ l- a-nsTO-rm-al-l-o-n~ a-n-a~ sol-l~ m-o-l-Uu-re~ a-cc-o-u-nll--~-g~ Over~ o-m--p-l-ex~ :::::::::::::::::: :::::::: ::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ''''''-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-~-:-:-:- :-:-'-:-:-:-:-:-:-:-:*:-:-:-'-:-:-:-:-:d-:-l:-'-:-:-:-'-:-:-:-:-:~-:-:~:-:~:-:~-~-: -~:-:-:-:-:-:-~:-:1---:-:-:-----:-:-:-:-:---:-----:- - -:-:---:-:~-:---:-:1-1:-:-:-:-:-:---:-:-:-:-: ~-I-~-~.~ Io lne I~V.V3Kh~ lne ral~rall ru-n-orr~ lra-n-d -orm-all-~-n~ * .-as~ ed 1l-ma-leo~ ~-ad ed ~ o-n~ a-n-lec-ec ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ 1) a h di h d ad i f il i d h fff f i i i b ................................................................................................................................ .. ' ' ' ' ' '-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-'-'-:-:-:-:-:-'-:-:-:-'-:-:-'-:-:-:-'-:-:-~:- - -:-:-'-:-:-:-'-:-:-:-:-:~-:-I:-'-:-:-:-'-:-:-:-'-:-'-:-:-:-:-:-'-'-:-:-'-:-:-:-:-'-'-:-:-:-:-:-:-:-'-:-'-:-:-:-:-:-:-I:-:-'-:-:d-:-~:-'-:-:-:-:-'-:-:-: re-p-rese-n-l-l-n-g~ l-n-e-m~ as~ c-u--m-u--l-all-d -e~ p-re-cl--p-l-lal~-o-n~ * .-lln~ a~ l-l--n-ea-~ ~-e-cay~ o.v..-er---ll-m-e~ u-lTre-re-n-l~ Kh--~d ~ n-a-a~ ~-eve-l-o ::::::::::::::::::::::::: :::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ............................................................................................................................. '"""""""""""""""""''''''"""""'f"""h"""'' "d 1"""""'''' d ''''" ''' ........................................................................................... , .............. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:-:---:-----:-:---:-:-:-:---1:-:---:-:1-:-:-:-:-:-:-:-:-:-:---:-:---:-:-:-:-:---:-:-:-:-:-:---:---:-:-:---:-:-:---:-~:-:1-:-:-:---:-:-:-:---:---:-:-:-:---:-:-:-:---:-:~-:-:-:-1:---:-:-:---:-:---:-:-:---:-:---- 1 :-: mO-d -e-l-d ~ w.em. m-osily. e---venl. oa-d e-d ~ a-e-d ~ In-e. l--n-lerd lo-~-m. perl-oa. w.-ad ~ n-ol. ex-p-l-l-c-l-lly. m-oa-e~-ea. ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::: '"""""""""""""""""""""""""""' 'h" ""' '' '''' ' ' ""' WS"' ' 'S""' '1'1" ' i" ' ' ' "'' " '' " ""' f"'' h'' ""' " ' 'b'l" " '' "" f"''' 'i'' a'l"l' '' "' '~"" ' ' '' '~''' '' ' i'' " ''""'' 'i"' ' ""i' ""' " ' ' ""' ""' '' 'd'' "1'' ''" -:-:-:-:-:-:-:-:-:-:-:-:-:~-: 1 :---:-:-----:-~:-:-:-:-:---:-:1-~-:-:---:-:-:-:-:-:---:---:-:-:-:~-:-1:---:-:-:---:-:-:-:-:-:-:-:-:---:-:---:-:-:-:-:-:-~-:-:---:-:-:-:-:-:---:~:---:-:-----:-:-:---:-:-:---:-:-:---:-:-:-:-~:-:1---:-:-:---:-:-: ln-al~ a-l-~-owd ~ l-n-e~ user---lo~ c-n-a-n-g-e~ l-n-e~ p-ro--e-a-u--ms~ as~ need ~ loo-l-s~ a-e-d ~ lec--n-n-l-q--u-e$~ ~-eco-m-e~ ad -a~-l-a-~-l-e~ ~econ-d ~ in-e :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:-:---:-----:-:---:-:-:-----~---:-:. :-:---:-:-:-----:-:-:-:~-:-:---:-:-~:-:-:---:-:-:-:-:---:-:-:---:-:~-:~ :-:---:-:-:-:---:-:-:-:-~:-:---:-:-:-:-:---:-:-:---:-:-:---:-:-:---~-:-:---:-:---:1-:-:-:-1:---:-:-:---:-: m-a-cn-l--n-e. l-nle-ra-~l-o-n. l-s. n-ee-a-ea. D-eca-u$-e. or in-e. n-alu--re. oT. nyd -ro-l-~-g-l-c. ro-reca-~. p-ro-~-l-e-m-d. Kunorr. p-ro-c-ess~- ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: '"""""""""""""""""' ' '' " ' "' ' ' h' """ " ""'d'-ffi'' '' '1' ""'' ' """' ' '' ' ' " ' """i"' """' i'' "' '1'' """' '' 'd'' '1" '"""" i'' f' '1"1' ' ' -'' '' n""" '' d'' '1" ""'' h'' ' - '' ""'' '' ' """h"i'' 'h'l" ""'' ' " ' '' " ' ' '' i' ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ''''''-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-'-:-:-:-:-:-:-:~-:-:-:-:-'-:-:-:-:-:-:-'-:-:-'-:-'-:d-:-:-:-'-:-:-:-'-:-:-:-:- - -'-:-:-:-:-'-:-:-:-:-:-:-'-'-:-'-'-:-l: d:l-'-'-:-:-:-:-:-'-:-:*:- - -:-:-:-:-:-:-'-:-:-:-:-:-:-:-:-'-'-:-'-:-'-:-:-'-:-: re-q-u--l-~ exle-~-d I* e~ ca-l--l-~-rall-on~ e--d en~ o-n~ a~ seaso-n-a-l~ Das-l-d ~ hu~nermore pmlpilallon Ood e- all-o-~-d ~ a-n 2"""""""""""""""""2''""'""" dj :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ' ' ' ' ' '-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-:d-:-l:-'-:-:-:-'-:-:-:-:-:-'-'-:-'-:-:-:-'-:-:-:-'-~-:-:-'-:-l:-:-:-'-:-:-:-:-:~ :*:-~-'-:-:-:-:-:-'-:I- - -~-:-'-:-:-:-'-:-:-'-:-:-:-'-:-:d-:-:-:-:-:-'-:-:-:-'-:-:-:-'-:-'-'-:-:-:-'-:-'-: u-se. in-e. m-o-a-e-l. d lln amerenl parameler cnolced ano lnpul d equences lo aelermine lne error slfud lure or lne ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::::::: '"""""""""""""""""""""""""""' 'h" """ ' 'i" f' '1"1' ' ' '-' -' ff""' ''' ' ' d"' '' "'' h" ' ""' " ' ' ""' ' 'C' "" i"l'l"""' " ' ""i' ""'' ' '' '' i"' '' " '' ' ""' ' '1""" ' 'i' ' ' '-' ' '-'-' -' ' ' '' '-' -' i'' -' ""'~S" ' )''""' h" ' ' "" -:-:-:-:-:-:-:-:-:-:-:-:-:-:---:-----:-:---:-:-:---:-:-:-:-:-:-:-:-:-:---:-:---:-:-:---:-----:-:---:-:-:-:---:-:~-:-:---:-:-:-:-----:-:-:-:-:---:-:-:-:-:---:-:-:---:-:-:---:-:~-: d:-:---:-:1-~---:-:-:-:-:-:1-:-:-:-:-:-:---:-:---:-:-:---:-: m-eas-u-mm-enls~ a-re~ $-pa-l-l-a-l-ly~ l-u--m-p-ea~ re-p-resenlall-ons~ or~ l-n-e~ s-o-l-l~ -o-l--u-m--n~ as~ u--p-per~ a-n-a~ l-od .-e-~ slo-rag-e~ zo-n-es~ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::: ::::::::::::::::: :: ..................................... -- -- ' '' " ' '' """ ' " " ' ' 'I""'' h" "" ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::: :-:-:-:-:-:-:-:-:-:-:-:-:---:-:-:-:---':-~:-:1-----:-:-:---:-:---:-----:-:-----:-:---':-:-----:-:-----:-:-:-:-~:-:'-:-:---------:-:-----:-:-:-:-----:-----:-----:-:1-:-----:-:-----:-:-:-:-:-~1-:-:---:-:-----:-:---- 1 --:-:---:-:-:*:-1:---:-:-: lln va~lOg Ilme s al~ ~learly Ine callorailon or lnese moo ls ls a eniral lasK ror lne KL~. ::::::::::::::::::::::::::::::::::::::::::::: :::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :::::: ."""""""""""""""""""""""""""'0'' " ' "'' h" ""' ' 'I'i'b" ' ' ' 'd""' ' 'i"' f.a.l'l" ' " " '' ff""' '' 'd'' 'I""'h" ' ""b" ' ' """ ' " d""' ' ""d" ' ' ''' '-'' '' "" h" ""' ''' ' i"' i' ' ' i' ' ""' '' ' ' ' """"' " "i" ""h'' d" ' ' ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ''''''-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-'-:-:-:-'-I:-:-:-:d-:-:-:-'-:-:-:-:-:-:-'-:-:-'-:-~:-:~-:-:-'-:-:-'-:-:-'-'-:-~:-:-:-'-:-:-:-:-~:-:l-'-:-:-:-'-'-:-:-:-:l-'-:-:-:-'-'-:-l:-'-:-:-'-'-:-'-:-:-:. :-:-'-:-:-:-'-:-:-:-:-:-'-:-:~-:-:-: u-sea~ lo~ esll-m-ale~ ln-e~ De-n-a* lor or slreams or. ood -l-cd ~ o.T~ waler~ al~ l-n-e~ m-o-re~ ln-a-n~ ~ ~-~ romcasl~ l-o-call-~-n$ ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 2"""""""""""""""""2P''' i''""l :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: I , , , . ~ :1: - ,.,,,, ~, . ~ f . A. d ^.~.:_.~:.:-:-:-:-:d-:-1:-'-:-:-:-'-:-:-:-:':' ':':':':' ':'~' ' ' ' ' ' ' ' ' mO-d -e-l--l-n-g. l-nio. in-e. l~-vv-~-K-~-~. Ine enilre nyar.ograpn ls usea ln Torecasn. ca-l-l-~-rall-o-n. a-~-a. ve-rlT '"""""""""""""""""' ' '1"' '' '' '~ ' h ""'' h '' "' " 'I"""h'' ' ""'b' " " -""'' ' ' i"' '' ' ' d"""""' h'' ~ m-- - S"""' ' " " a'' ~ ' 'b' ' ""' f""' ' ' i"' ~ ' " i"' " ' i"' '' '1" 'd'i"' " """ :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: .""""'"""""""""""'Wa' 'e'-'-Ope'' ati~ al""'M' del""'t'~OPER)""""'FI'' ' 'd-'-Wa' e""'M'' d'e'l""(F'LD~""""a' d"""Da' ""'B" ea'k""M'' d-' I""'~'DAM'B'R~' t' ""'' a- a-- -~e~ '"""""""""""""""""'i' ' """ ''' ' ' d" ''' ' ""~' ""' d'' ' ' ' i'' '' """ h'' ""'h'' d" " ''' " 'h"""d'' ' " " ' " ' " "'""'' 'h'' """' '' d" I"i'' " ""'' f""' i' ' ''""'h' d" ' '' 'I'i' """"'fl" ' 'd""' ' ' ' """ ' " " i"' " '~ .a -:-:-:-:-:-:-:-:-:-:-:-:-:-:---1:-:---:-:-:---:-:---:-:-:-:1---:-:-:---:-:---:-:-----:-1:---:-:-:-:---:-:-:---:-:---:-:-:- - -:-:---:---:-:-:-:---:-:-:---:-:-:---:~-:-:---:-:-:---:-:-~:-:-:-:-:---:-:-:---:-:-:----1-:-:-:-:---:-:---:-:-:-:-- am omaK are Im po~anI an a sopn lsil(awo componenn Or DoLn TIasn rloo an a mal n siem river rlooo na~a a mll ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: T-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-.-. -.-:1---:-:-:---:-:-:-:-:~-:-:~:-:~:-:~-~-:_-~:-:-:-:-:-:-:-:---:-:-:-:---:-:-:-:---:-:-:-:-:-:-:-:-:---:---:-:-:---:-:-:-:-:-:---:-:-:-:-:~-:-:---:-:-:---:1-:---:---:-:-:-:-~:-:1---:-:-:- n-e~ l~-v.v-~-K-~-~ sewed ~ m-o-re~ las-~ l-n-a-n~ l-~--~ a-e-d ~ ca-l-l--~-ra-l-l-~-~ a-l-~-~-e~ ll~ ls~ a-l-d o~ a-n~ e ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: 2"""""""""""""""""2' ' ' h " '' " """' :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ~ ~ * q-u-a-nlllall-ve~ p-rec-l--p-llall-on~ To-reca-Ud ~ a-n-a~ q-u-a-nlllall-v.-e~ le-m-p-e-ralu-~ ro-mca-Ud ~ rro-m~ vv-~-ws~ n-ea-d -w.-aler---g--u-l-d -a-n-ce~ a-n-a~ T-I-as-n~ ::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::: :::::::::::::::: ::::::::::::::::::::::::::::::::::::::::::::::::: fl - d id ~ w O ~ d da --- h - -- i h i 1 d i 1 - - )--- h -- WS- -S--i -- 1 - h -- ..................................... ......................................................................................... :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: -:-:-:-:-:-:-:-:-:-:-:-:-:~-:~ :-:---:-:-:-:---:-:---:-:-:-:-:---:-:---:-1:-:---:-:*:-:-:---:-:----1-:-:-:-~:-:-:---:-:-:-:-~:-:1---:-:-:-----:-:-:~:-:~:-:/::-----:-~:-:-:---:-:-:-:-:-:-~-:-:---:-:---:-:-:---:-:-:-:-:-:-:---:-:---:-:-:---: ll-o-n-d ~ re-l-alea~ lO~ ln-e~ vv-aler~ K~-o-u-~-es~ ~-~-recasl-l-n-g~ ~le-m~ a-n-~ ln-e~ ~xle-n--d ec~ ~lre-a-mrl-od ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: .......................................................................................................................................................................

OCR for page 19
26 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES excitement among the RFC staff. The system allows great flexibility to manipulate observations and guidance. It also allows the forecaster to use different components effectively and efficiently. Once the observation and guidance fields (e.g., NEXRAD and QPF fields) have been integrated into the database, the NWSRFS aids the forecaster with three major tasks. First, it uses a rainfall-runoff model to determine the fraction of the precipitation that is partitioned into infiltration and runoff. This step includes estimating the snow accumulation and its subsequent ablation. Second, it applies models that estimate the time required for the runoff to reach the river channel and flow past the forecast location. Finally, the runoff vol- ume is tracked as it moves downstream within the channel. As part of this third stage of hydrologic forecast preparation, changes in the peak and spread of the flood wave are esti- mated by "river-routing" procedures. Although the NWSRFS represents a timely and signifi- cant advance over the methods it replaces, the committee notes a serious deficiency in the system as currently config- ured. The science that underpins the applications lags the advanced technology considerably. The Hydrologic Re- search Laboratory and RFCs have redirected extensive hu- man resources and specialized staff efforts from hydrology into systems development. Although there has been a con- siderable effort directed at coding and systems building, only limited resources have been directed toward expanding the scientific foundation underlying the model. Conducting tech- nical exchanges with the USGS on modeling topics is a posi- tive step for the NWS in improving the scientific basis for its hydrologic models. Conclusion. A solid, technological base has been established for the NWSRFS; but because the underlying hydrologic science is out of date, the committee concludes that the NWSRFS as currently configured may not be capable of meeting future needs for improved hydrologic services. The NWSRFS should be reviewed to determine where more re- cent models and data sources might be incorporated into the strong technology base. For example, digital precipitation estimates produced by NEXRAD with high spatial resolution indeed represent a breakthrough for hydrologic operations that heretofore re- lied on point rain gauge observations. Furthermore, high- resolution (3-arcsec, or less than 90-m) digital elevation data are now available for the entire United States. Hydrologic models based on these elevation data provide unprecedented capabilities to characterize the landscape and drainage net- works. Spatially distributed, continuous simulation hydro- logic modeling4 is now possible (Krzysztofowicz, 1995), yet it has received minimal attention in the NWS. Instead the 4Spatially distributed, continuous simulation hydrologic modeling is an approach to modeling that uses detailed, digital precipitation and topogra- phy data for collective analysis. emphasis has been on the calibration of soil moisture account- ing models, a spatially lumped procedure for smaller water- shed subdivisions. Such lumped parameter models were de- veloped in the late 1960s and early 1970s when neither high- resolution precipitation data nor elevation data were avail- able. Moreover, the limited computational capabilities at the time restricted the development of advanced models. As another example, the current NWSRFS uses the "unit hydrograph"5 methodology to route the precipitation that contributes to all components of the runoff (precipitation excess) over hillslopes into drainage networks. A unit hydrograph is the hydrograph associated with a unit of pre- cipitation excess; depending on the amount and duration of precipitation excess, the unit hydrograph is scaled to pro- duce the hydrograph at a given forecast location. The unit hydrograph essentially captures the drainage pattern and tim- ing over the watershed drainage through the forecast loca- tion. It is an empirical tool developed decades ago in the absence of detailed topographic characterizations of drain- age basins. Unit hydrographs also make an invalid assump- tion (one that was the best available decades ago) about the response of the basin to differing amounts of precipitation excess; that is, they assume a linear response in the behavior of the watershed. Science and technology are two sides of the same coin. The technology that supports scientific advances must be in place before the new science can be used. If the science underpinning the technology remains static, the full poten- tial of the technology cannot be realized. Conclusion. It is critically important that the scientific pro- cedures used in the NWSRFS receive increased attention and that systems development not be allowed to become an end in itself. Recommendation 3-10. The Office of Hydrology should place greater emphasis on building the scientific foundation for the NWSRFS and integrating it with the existing strong technology base. In particular, it should consider using spa- tially distributed, continuous simulation hydrologic models to replace and/or augment, spatially lumped and parametric models (i.e., with heavy reliance on parameter calibration rather than physical principles to predict hydrologic condi- tions) in the modernized NWSRFS. Furthermore, the incor- poration of empirical unit hydrographs into the NWSRFS should be reconsidered in light of the detailed and distrib- uted digital precipitation and topography data that are now available. In each RFC the levels of expertise in the use of NWSRFS calibration procedures vary widely. Life-cycle support for the NWSRFS software is also necessary. This includes software 5A hydrograph is a continuous plot of instantaneous discharge, versus time, at a point along a river or stream. It results from a combination of factors related to the topography, land use, geology and, most important, storm precipitation over the area drained through the river or stream.

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES trouble shooting, interactions with the users, and integration of new procedures and technologies into the software. Ad- equate training is clearly an issue. Conclusion. To take optimal advantage of the NWSRFS potential, perhaps the most important need is for advanced training for forecasters in the use of both the calibration fea- tures and the interactive capabilities of the system. Recommendation 3-11. To improve consistency in the use of the NWSRFS among RFCs and within RFCs, systematic oversight and effective training programs should accompany the installation of NWSRFS systems in RFCs. Training should be provided for all appropriate RFC staff to ensure sustained proficiency in the calibration, verification, devel- opment, and use of the NWSRFS. Flash Flood Guidance The NWSRFS provides hydrographic forecasts at a lim- ited number of forecast sites within each RFC's large area of responsibility. These sites generally are located on large streams and rivers that have long lead times for flood peaks. The more short-fused flash flood hazard occurs on a consid- erably smaller scale that is more compatible with WFO hy- drologic service areas (HSAs). To assist with the WFO flash flood program, the NWSRFS produces flash flood guidance products, which document the precipitation thresholds re- quired to initiate flooding. Forecasters in WFOs use this guidance in conjunction with local application software to help produce flood and flash flood warnings for their HSA. Flash floods generally result from a few hours of heavy rainfall. However, many conditions can contribute to such events. The break up of ice jams and the failure of hydraulic structures (e.g., dams and levees) are also capable of creat- ing flash flood conditions. Other physical factors that affect flash flooding include rain intensity, antecedent soil mois- ture conditions, rainfall type, urbanization and extensive impervious surfaces, steep land slopes, and conditions that produce rapid snowmelt (rain on snow, warm humid air, va- por condensation, etc.~. The WFO forecaster issues flash flood warnings and watches by comparing the observed or forecasted precipita- tion amounts with the guidance values present in the flash flood guidance. This task is performed with the WFO Hy- drologic Forecast System (WHFS) application (see next sec- tion), which incorporates "threshold runoff" values.6 At the WFO, forecasters may use two subsystems of the WHFS to deal with flash floods. The Site-Specific Hydro- logic Predictor System (SSHPS) works much like the NWSRFS, in that the hydrograph at a specific point in the stream represents the forecast. Headwater guidance from RFCs, along with independent calibration and soil moisture accounting, prepares the SSHPS system to predict conditions 6Threshold runoff values are the amount of runoff required to bring streams and channels to a full bank condition. 27 at points within the HSA. The WFO forecaster uses the sec- ond WHFS subsystem, the Area-Wide Hydrologic Predic- tion System, in conjunction with flash flood guidance and threshold runoff applications, to issue flash flood warnings and watches for areal zones within the HSA. Conclusion. The WHFS, with its capabilities for both site- specific and area-wide modeling of the flash flood hazard, has great potential for dealing with the zero to six-hour flood problem if technical and scientific, training, and operational procedures problems are resolved. Whereas the flash flood guidance is produced by soil moisture accounting with a spatially lumped model, thresh- old runoff is estimated based on limited data for the digital elevation model, land use, and river reach (Georgakakos, 1986, 1992~. Flash flood warnings and watches are therefore produced by a mixture of lumped and distributed models. Conclusion. The scientific foundations of both flash flood guidance and threshold runoff are derived from decades-old techniques that need significant revisions. The detailed com- ponents of both guidance products also need extensive testing as soon as is practicable before and during implementation. Recommendation 3-12. The NWS should improve the sci- entific basis that underpins the forecasting of floods that oc- cur in the zero to six-hour time frame. WFO and RFC staff should be enabled to contribute to this effort by facilitating their access to adequate training, continuing education, and university cooperative programs. Furthermore, the staff should be able to access state-of-the-art geographic informa- tion systems, databases that include digital elevation, drain- age, and land-use data for use in NWS models. Apart from purely technical issues, other issues need to be considered when flash flood warnings and watches are generated. WFO HSA boundaries may intersect RFC bor- ders. Guidelines for the generation of flash flood guidance and its adjustment at WFOs that maintain consistency be- tween neighboring regions have not been defined. Conclusion. Because there is no provision to ensure the con- sistency of areal flash flood guidance generated from adja- cent RFCs, abrupt changes in threshold runoff at RFC bound- aries may result (Sweeny, 1992~. Recommendation 3-13. The NWS should develop guide- lines to ensure consistency in RFC calculations and WFO- specific adjustments of area-wide flash flood guidance. The NWS also should clearly define the roles and responsibili- ties and the expectations for the support of site-specific flash flood forecasting at WFOs. Weather Forecast Office Hydrologic Forecast System The WHFS software application represents the soft- ware environment in which WFO forecasters and service

OCR for page 19
28 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES hydrologists manage hydrometeorological data, communi- cate with RFCs, model hydrologic conditions within the HSA, and generate and communicate hydrologic and hydrometeoro- logical products for the user community. As in the case of the NWSRFS, the Office of Hydrology has taken the lead in creating this application ahead of other AWIPS develop- ments. The evolution of the WHFS has also benefited from the involvement of the Service Hydrologist Working Group.7 Conclusion. WHFS development at the Office of Hydrol- ogy and the interaction with the Service Hydrologist Work- ing Group represent substantial steps toward achieving an integrated hydrometeorological system to forecast floods that have short lead times. Similar to the NWSRFS, the WHFS is a software system with an interactive graphical user interface. The distinguish- ing feature of the WHFS is that it is designed to enable WFO meteorologists (as opposed to hydrologists) to handle hy- drologic hazards within the HSA in an effective and skillful manner. WFO forecasters will have had some basic hydro- meteorological training and will receive assistance when the service hydrologists are available. In addition, RFC staff and the NWSRFS also influence the operational use of the WHFS system. WFO meteorologists are expected to use the RFC guid- ance and the WHFS capabilities to produce flash flood and site-specific flood forecasts. They are also responsible for ensuring the availability of warning and watch products to the user community in their HSA. These responsibilities are in addition to their routine warning duties for weather haz- ards, which entail many user products. The effective use of the WHFS (especially of the SSHPS) during periods of se- vere weather when flooding is a hazard will add an unusual burden to the WFO forecaster's workload. Conclusion. There is a potential danger that WFO forecast- ers may process RFC guidance products through the WHFS and issue hydrologic warning and watch products without taking adequate time for the careful scrutiny of RFC guid- ance. This matter is an especially acute problem in basins where hydraulic structures, small-scale land-use patterns, urban surfaces. etc.. complicate the local hvdrolo~ic re sponses to excessive precipitation or snowmelt. in such cir- cumstances, the hydrologic hazards of relevance to the HSA could receive inadequate attention despite the availability of high-resolution NEXRAD precipitation estimates. One es- sential key to the proper use of the WHFS is adequate train- ing for service hydrologists and WFO forecasters. 7 The Service Hydrologist Working Group is an advisory team, formed in July 1994, with members drawn from various field offices arid headquar- ters staff. The working group periodically meets with regional arid head- quarters research and development staff to discuss issues of concern to the field service hydrologists, which may affect the evolution of the WHFS and related AWIPS software builds. Recommendation 3-14. The NWS should reevaluate the staffing needs of WFOs with regard to their hydrologic re- sponsibilities. The number of service hydrologists should be increased so that each WFO has a program leader for WFO hydrologic operations, at least for the first year or two fol- lowing implementation of the AWIPS at each field office. The full capabilities of the WHFS can be realized only after WFO staff are adequately prepared to deal with hydro- logic forecasting during flood-water crises and severe weather conditions. Conclusion. The WHFS needs to be tested in the most chal- lenging of operational environments, for example, during complex, severe weather situations that create intensive fore- cast and warning workloads. Recommendation 3-15. Guidelines and procedures should be in place to ensure that hydrologic and hydrometeorologi- cal forecasts meet NWS requirements (e.g., for accuracy and timeliness) under even the most challenging of operational circumstances. Operational tests should be performed to con- firm that these requirements are met. Recommendation 3-16. As WFOs and RFCs complete the transition to modernized operations, their performance should be monitored across different geographic regimes (e.g., variable climatic and hydrologic conditions) as part of the existing but expanded risk-reduction demonstration and operational test and evaluation activities. Extensive opera- tional test and performance evaluation are especially needed for WFO functions that deal with hydrologic hazards that accompany other forms of severe weather within the HSA. The WHFS also needs more robust quality assurance pro- cedures for data. Such procedures will reduce the number of situations in which erroneous hydrologic forecasts are gen- erated based on faulty observations. Both software features and user training are required to minimize this problem. Advanced Weather Interactive Processing System A distinguishing feature of the NWS modernization of hydrologic operations is the initiative to develop much of the AWIPS applications software in house. Consequently the NWSRFS and the WHFS interactive forecast applications are among the first components of the AWIPS to be deliv- ered to NWS field offices. As a result of the development of the AWIPS applications inhouse, many key personnel in the Hydrologic Research Laboratory and RFCs have been dedi- cated to the tasks of coding and programming, thereby strain- ing already limited resources. The Office of Hydrology has provided a minimal baseline of pre-AWIPS equipment to each RFC. This equipment has been secured through a variety of means, including alloca- tion of government development platforms, local initiatives, and pre-AWIPS prototyping tasks.

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES Conclusion. The AWIPS is an absolutely essential compo- nent of the NWS modernization. The software development effort at the Office of Hydrology and at RFCs has led the NWS transition into a pre-AWIPS era. However, the acqui- sition process and the reliance on interim solutions in ad- vance of the full availability of the AWIPS might cause prob- lems for future NWS operations in hydrology. The variety of interim solutions could lead to difficulties in the maintenance of operational consistency and the migration of software to the AWIPS. Recommendation 3-17. The NWS should ensure consis- tency in the hardware acquisition process for RFCs and es- tablish guidelines for consistency in the migration to the AWIPS from existing interim solutions. Issues related to the portability of the applications software that has been devel- oped should be given high-priority attention. Advanced Hydrologic Prediction System As a broad suite of versatile forecasting tools, the Ad- vanced Hydrologic Prediction System (AMPS) has been de- veloped in response to national needs in water resources management. Water allocation under competing and in- creased demands (e.g., fisheries, irrigation, hydropower, and municipal use) requires higher-quality forecasts. These fore- casts are also required to have longer lead times than those required for flood watches and warnings. Furthermore, both low-flow and flood conditions need to be forecast with greater precision to improve water management activities. In support of long lead time and general (comprehensive) water resources management issues, the AHPS has tasks re- lated to the overall NWS modernization and to the NOAA operational infrastructure (e.g., forecasting storm surge con- ditions on inland lakes and coastal zones, supporting activi- ties related to in situ and remote data acquisition, and devel- oping hydrologic coupling to atmospheric models). The AHPS also supports partnership programs with other agen- cies that deal with water management. This is one of the major themes in the AMPS: a group of elements collec- tively known as the Water Resources Forecasting System (WARFS). The WARFS is an integrated modeling, data management, and analysis program in support of compre- hensive hydrologic services within the NWS. Its infrastruc- ture includes the NWSRFS and the NOAA Hydrologic Data System,8 in addition to other observation, modeling, and analysis systems. Long lead-time forecasts for sustainable development and efficient management of water resources are among its main objectives. One of the key features of the WARFS is the extended stream-flow prediction (ESP) program, which involves en- semble forecasting of stream-flow with long lead times. In The NOAA Hydrologic Data System will provide the integrated data management and analysis capabilities required by the AMPS. 29 the ESP, the NWSRFS, with its current values of state vari- ables, is integrated with different time-series traces of rel- evant historical precipitation. Clearly, a well-calibrated NWSRFS system is vital to the success of ensemble fore- casts, because the "open-loop" behavior of the models will strongly affect the long-lead forecasts. Each trace is weighted according to forecast conditions or occurrences of histori- cally similar climatological settings. The advantage of the ESP is that it provides probabilistic long-lead forecast prod- ucts that have important applications in water supply, flood outlook, and drought analyses. It also has the capability to deal with forecasts of minimum stream flow, which are used in the management of river navigation corridors and wildlife (e.g., fish) habitat. Conclusion. The WARFS objectives to provide long-lead and probabilistic forecasts of low-flow and flood conditions will be important additions to the suite of products and ser- vices produced by the modernized NWS. The ensemble fore- casting method used in the ESP is an innovative develop- ment in support of the WARFS. The main shortcoming of the WARFS and ESP components of the AHPS is the appar- ent lack of involvement by NWS field personnel and the user community during the program development. The suite of products and services should be driven by specific external-user requirements for their particular re- gional environments. The committee found that the WARFS and its products and services have been, instead, principally designed in isolation at the NWS Office of Hydrology. NWS field personnel who are in routine contact with their user community, and who will ultimately be responsible for gen- erating WARFS products and services, are well positioned to guide the design of these programs. Their responses to the committee's questionnaire (see Appendix) and comments made during committee visits to field offices indicated a lack of detailed knowledge and involvement in the program defi- nition. Recommendation 3-18. Field personnel and users of prod- ucts and services should have greater involvement in the fur- ther definition and development of the WARFS and other components of the AMPS. OPERATIONS Recognizing the complexity of and urgent demand for hydrologic forecasts to meet growing societal needs, the NWS has chosen to integrate its operational missions in 9"0pen-loop" behavior refers to the use of a model to simulate long-range conditions without any correction of its forecasts (through up- dating the model-state variables and parameters). Such long lead-time fore- casts without occasional corrections tend to reflect more the inherent char- acteristics of the model and less the content of observed (initial) values. As a result, the model' s robustness and realism are important to the accuracy of long lead-time forecasts.

OCR for page 19
30 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES meteorology and hydrology. The new hydrometeorological focus in a modernized NWS should enable the agency to implement improved hydrometeorological operations in a more efficient and effective fashion. To achieve that benefit, it is necessary for the NWS to blend the responsibilities of WFOs and RFCs. Conclusion. The NWS can best exploit the opportunity that the modernization affords not only by emphasizing technol- ogy but also by capitalizing on the overlapping aspects of hydrologic and meteorological science and technology and by developing new operations to aid the interaction and transfer of information between hydrologists and meteorolo- gists. Emphasis needs to be placed on the tools and scientific principles common to the meteorology and hydrology disci- plines in the NWS, while, at the same time, recognizing the distinctive nature and requirements of each. Weather Forecast Office and River Forecast Center Interactions With WFOs and RFCs now more than ever needing to function as a team, each has been assigned responsibilities designed to enhance and help meet their joint mission. This means WFOs will undergo a "cultural shift," with the entire forecast staff becoming involved in hydrology issues (whereas previously only the service hydrologists were in- volved in hydrology issues). At RFCs, an increased sensitiv- ity to and understanding of WFO and other end-user needs are the bases for extended hours of operation and for the increased frequency of updating hydrologic guidance and forecasts. A fundamental tenet that underpins the modernization of hydrologic operations within the NWS is the recognition that WFO service hydrologists are program leaders within their offices, not operational forecasters. However, most WFO weather forecasters currently do not perceive hydrology as part of their operational duties. Furthermore, during severe weather conditions when WFO forecasters primarily are con- cerned with generating warning products, the use of interac- tive hydrologic forecast programs might receive less atten- tion than is warranted. This means that flash flood and head- water guidance from RFCs could be used without adjusting for locally known conditions within the USA. Moreover, local duties at WFOs and RFCs are event driven; they are not program based (i.e., hydrologic policies versus meteorological policies). This situation has led field personnel to express concerns about the ability of NWS man- agement to integrate successfully hydrology and meteorol- ogy into the operational environment at WFOs. Fortunately the hydrometeorological analysis and support (HAS) unit at each RFC represents a unique opportunity to overcome the historical separations that have existed in field operations between hydrology and meteorology personnel. Indeed, the position announcements for HAS staff indicate a primary responsibility for leading the effort to improve the interac- tion between WFOs and RFCs an effort that will be facili- tated further by the fact that many meteorologists are being assigned to the HAS units. Nevertheless, training and staffing are fundamental issues that are implicit in the modernized operations of a WFO and a RFC. Modernized equipment exists, along with new mod- els and algorithms, and a new organizational structure will soon be in place. However, sufficient staff must also be in place who have the appropriate training to take advantage of these modernized capabilities. As noted in Chapter 1, only 80 of the 119 WFOs will have an assigned service hydrolo- gist. Moreover, today's WFO forecaster will face a myriad of complex hydrometeorological issues that are not easily dealt with especially in complex basins with physical con- trols such as reservoirs and dams. Conclusion. Even with the support of HAS personnel at RFCs and service hydrologists at WFOs, hydrometeorologi- cal forecasting at WFOs may at times produce an excessive workload for the planned staffing. Responsibilities and staffing levels may have to be rethought and adjusted over time. Nevertheless, appropriate training will be crucial (see Chapter 4~. Another concern is that river basins have been appor- tioned among the various WFOs according to geopolitical boundaries rather than physiographic boundaries.l This ap- portionment will cause confusion among those users who do not know where to acquire hydrologic forecasts and warn- ings for their area. NWS field personnel are also concerned about their ability to maintain continuity and consistency in hydrologic forecast and guidance products because hydro- logic service areas of responsibility are no longer coinci- dent with county warning areas of responsibility. For ex- ample, four different RFCs have responsibility for parts of Colorado (e.g., the Denver WFO receives guidance from three RFCs). Conclusion. To avoid these problems, it will be incumbent on NWS management to ensure the continuity and consistency of hydrologic products issued or used by RFCs and WFOs in overlapping areas of hydrologic service responsibility. Other operational issues of concern to field personnel in- clude questions about how WFOs that use site-specific hy- drologic models in the WHFS will be able to reconcile their results (especially at collocated forecast points) with those produced independently at RFCs. What if the two sets of forecast results disagree, especially during a short-fused emergency situation? What established protocols are in place to resolve differences that might arise between the WFO ser- vice hydrologist and RFC hydrologic forecasters? Although iGeopolitical boundaries such as county warning areas are assigned to WFOs based on the severe weather detection coverage provided by the nearest NEXRAD.

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES the WFO is expected to coordinate any changes to RFC- issued forecasts directly with the issuing RFC, the question at the WFO is not one of revising RFC guidance, but adapt- ing RFC guidance. Conclusion. In the situation of an evolving hybrid event that is intermediate between a flash flood and a river flood, NWS policies do not make clear who has the final word in regard to the forecast and warning products actually issued to the public. The committee feels that WFOs should have the ba- sic responsibility for continuity and consistency of flood products going to the users who receive such products from WFO communication links that operate 24 hours per day. Recommendation 3-19. The NWS should consider the need for more personnel in the hydrometeorological forecasting function. A formal "task analysis" of this function should be considered if difficulties are identified during operational test and evaluation and risk-reduction activities. Neverthe- less, adequate training and cross-training is vital for WFO and RFC staff with hydrometeorological forecasting respon- sibilities. Recommendation 3-20. LISA boundaries should, to the ex- tent practical, be divided along basin boundaries rather than geopolitical boundaries. In areas where adjacent HSAs are under the supervision of different regional offices, the NWS should ensure that the services of these offices are provided to users with the same operational philosophy and in the same format for the same product. Recommendation 3-21. NWS headquarters should develop clearly stated policies that help WFOs and RFCs reconcile apparent discontinuities in forecasts that occur when site- specific models at WFOs produce hydrologic warnings and forecasts that conflict with RFC guidance for the same fore- cast point. Data Archiving, Verification, and Quality Assurance A comprehensive data archiving system is the essential ingredient around which future improvements in the hydro- logic forecast system of the NWS must be built. The current system fails to archive and efficiently retrieve most of the data that will be needed for twenty-first century improve- ments in NWS hydrology. These data include: the hourly precipitation summaries from Stage II and Stage III processing of NEXRAD data rain gauge and stream gauge observations used to cali- brate NEXRAD precipitation patterns or to calibrate and validate hydrologic forecast models daily knowledge of basin characteristics such as veg- etation indices, antecedent moisture conditions, and major construction stage-discharge rating curves that relate flow volumes to stream gauge readings 31 records of previous stream flows documented on up-to- date E-l9 forms of the NWS The NOAA Hydrologic Data System, now under develop- ment (NWS, 1996b), represents an attempt to meet the needs of the Water Resources Forecasting System and of various climate programs. Although it has not yet been implemented, the NOAA Hydrologic Data System appears to meet the data archive needs of operational activities in river and flash flood forecasting. Conclusion. A lengthy archive of all basin data is essential to calibrate hydrologic models (old and new) and improve model performance during critical high-water and low- water situations. For example, hydrologic models have always suffered from a lack of knowledge about the distribution in space and time of precipitation that produced floods or led to drought conditions. Yet gauge-calibrated precipitation estimates from the NEXRAD network provide a unique opportunity to produce hydrologic guidance and forecasts with an unprec- edented level of detail (i.e., at the subcounty level and on time frames of one to three hours). Satellite imagery can provide details on the areal extent and intensity of precipitation. Conclusion. With increasing demands for hydrometeoro- logical data in an interactive forecast environment, efficient methods for accessing, storing, and viewing these data are required. Recommendation 3-22. The suite of precipitation products produced by the NEXRAD network, along with accompany- ing surface rain and stream gauge information, should be archived by the NWS for future use when new hydrologic models require calibration before they can be implemented. The NWS should ensure that appropriate access, storage, and visualization methods, such as those planned in the NOAA Hydrologic Data System, are developed or adapted for use with the entire spectrum of hydrologic data. A second essential ingredient to improve hydrologic ser- vices is an adequate forecast verification system. Such a sys- tem must provide a baseline that documents previous fore- cast skill levels and also detects small improvements in forecast skills that result from new models being developed, calibrated, and implemented. The NWS is pursuing such a system in its National Hydrologic Forecast Verification Pro- gram (NWS, 1996b). Conclusion. Currently the verification of hydrologic fore- cast products is inadequate. Hydrologic model development and the incorporation of new scientific tools into a modern- ized work environment should be accompanied by a rigor- ous verification program to document what progress has been achieved. The National Hydrologic Forecast Veri- fication Program represents an important set of plans for overcoming this deficiency in the hydrology program of the NWS.

OCR for page 19
32 ASSESSMENT OF HYDROLOGIC AND HYDROMETEOROLOGICAL OPERATIONS AND SERVICES Recommendation 3-23. The NWS should implement and provide the sustained support that is needed to continue the development and operation of the National Hydrologic Fore- cast Verification Program. Although a comprehensive data archive is essential to achieve future improvements in the hydrologic forecast sys- tem, the quality of data in that archive is at least as impor- tant. Nowhere is the phrase "garbage in, garbage out" more relevant than when applied to the data used to calibrate the performance of new or improved hydrologic models or to produce a real-time forecast during critical events. All surface-water hydrology models live and die by the quality and quantity of input precipitation patterns, state variables of the models, and stream-flow hydrographs. The Hydrology Panel found general agreement among NWS field office hydrologists that the quality assurance of precipitation and river gauge data from telemetered networks is severely lacking. Most quality assurance is achieved by manual methods and is limited primarily to "screening" data received from the cooperative observing network. The prob- lem becomes especially acute when the data are gathered by many different agencies at all hours of the day and night. Conclusion. Although the NOAA is establishing the NOAA Hydrologic Data System, this system likely will not in itself improve the quality of data used in real time at the local WFO and the RFC, which determine the spatial and tempo- ral scale at which forecasts are prepared. To alleviate this problem, the HAS units, service hydrologists, and hydro- meteorological technicians need to respond as vital compo- nents in the quality assurance effort. It is important to note that quality assurance is not limited to flagging extremes in the data stream. Instead, accepted practices in quality assur- ance require a mix of daily oversight that ranges from auto- mating the detection of data extremes to tabulating frequently missed observations. A considerable effort is needed in NWS field offices before these "low-priority duties" are elevated to their proper status. Recommendation 3-24. NWS management should take ag- gressive steps to implement data quality assurance proce- dures at the local level for archivable precipitation and river gauge data. These procedures should be automated to the extent feasible, deal with the full range of hydrometeoro- logical data, and follow accepted standards for quality assur- ance. The NWS should establish guidelines for consistent use of data quality assurance procedures at both WFOs and RFCs. These guidelines should be reinforced by appropriate training of personnel. Data providers from cooperative agen- cies should be invited to participate in training activities where possible. The full implementation of the AWIPS will significantly improve NWS capabilities for archiving, verification, and quality assurance of data. Data Source Reliability Data for use in hydrology and hydrometeorological analy- ses and models are the lifeblood of river forecasts and flash flood products. These data include information from auto- matic and manual sources such as stream gauges, precipita- tion gauges, cooperative observers, flood warning systems, satellite-relayed data, telemetered data and, most important, data from the NEXRAD network. Satellite precipitation estimates can contribute useful hydrologic information in some areas. It is possible to make an accurate forecast with primitive models and complete, accurate data. It is not possible to make an accurate forecast with inaccurate or incomplete data, even with the most advanced models and interactive tools. The NWS provides a nationwide river forecast service, with forecasts for 4,018 stream gauge locations. Of these 4,018 locations, 77 percent are operated and maintained by the USGS. Over the past two decades, NWSRFS require- ments have increased because of expanding U.S. population, urbanization, and economic growth. At the same time, as the NWS requirements for surface- water information have increased, the number of stream gauge stations has declined steadily over the last 10 to 15 years, primarily because of a general tightening of federal, state, and local agency budgets. Another major impact is on the historical archive of surface-water information used by the NWS. The archive of surface-water information plays a pivotal role in hydrologic model calibration, development, and refinement of hydro- logic forecast procedures. The decline in surface-water in- formation has resulted in a loss of continuity at many previ- ously archived stream gauge stations. These losses, although not apparent in the short term, will have long-term effects on such NOAA programs as the Water Resources Forecasting System and global climate programs such as the Global En- ergy and Water Cycle Experiment (GEWEX) and the GEWEX Continental-Scale International Project. The NWS hydrology program will require additional real-time and archived surface-water information to support current and future river forecast services for the nation. Conclusion. Interagency cooperative programs for hydro- logic data collection are critical to the success of river fore- cast and flash flood guidance forecast programs within the NWS. However, ownership of these data sources are distrib- uted across federal, state, and local government or private networks. Abrupt funding changes and uncertainties cause some data sources to be unstable, unreliable, or subject to short-notice curtailment or elimination, sometimes with no apparent coordination or consideration for the impact of their loss on NWS operations. It is essential that a shared owner- ship exist for the nation's water management and flood warn- ing infrastructure, lest this infrastructure lose its ability to meet growing operational demands.

OCR for page 19
MODERNIZATION OF HYDROLOGIC SERVICES Recommendation 3-25. The NWS, along with other federal agencies and local and state governments, should coordinate hydrologic and hydrometeorological data requirements, data collection, and processing. Priorities among these data should be set and appropriate funding allocated by the par- ties involved to maintain a consistent, reliable set of data for national and local flood forecasting programs. The NWS should exert leadership to forge an explicit partnership for sharing data collection resources. PRODUCTS AND SERVICES A service agency such as the NWS is known by the qual- ity, relevance, and ease of access and use of its products. Nowhere are these traits more critical than in river and flash flood situations where lives and property are threatened- situations that often arise late at night and in rural areas. In these situations, a forecast and warning system provides an unacceptable level of service when the forecasts and warn- ings are inaccurate or difficult to decipher, or when they fail to reach the population at risk. Repeated experiences dem- onstrate that the current dissemination technology often fails to bring critical information to those at risk. Too often, dis- semination is hampered by the slow process of individually contacting city officials and emergency managers (including law enforcement and public works agencies) in various mu- nicipalities within the USA. The potential consequences of this situation have become more severe because of the larger percentage of the U.S. population now living in flood-prone areas, the escalating cost of repairing flood damages and the high economic toll of such disasters in a complex economy, and the increased scrutiny that government agencies are undergoing during times of diminishing resources and budget crises. 33 Conclusion. To achieve the dividends of modernization re- quires that the NWS and its myriad of users understand each other' s needs and capabilities, establish highly efficient tele- communication linkages, and support or advocate mutually beneficial programs. In the committee' s view, the NWS must move from a vision of providing technology to a vision of providing new services based on opportunities to implement advances in hydrologic science and technology. Interaction with the user community is an ongoing activ- ity that spans the NWS. It needs continual nurturing. Yet NWS field hydrologists repeatedly voiced their concerns to the committee that not enough dialogue and feedback had occurred with external agencies. Several individuals felt that external users had only a minimal impact in defining new hydrologic services. At the same time, these field hydrolo- gists emphasized the importance of being in the pipeline of hydrologic information produced by other federal agencies so as to minimize miscommunication during critical, short- fused events. In a modernized WFO, service hydrologists and warning coordination meteorologists will have an im- pact on the coordination role with external users. Recommendation 3-26. The NWS should continue to work with the user community to determine community needs. In particular, the NWS should focus on user concerns that may develop in regions where political boundaries and basin boundaries do not coincide. Recommendation 3-27. The NWS should develop creative and affordable methods to disseminate its hydrometeorologi- cal information by means other than direct, individual con- tact (e.g., graphical warnings). These new dissemination methods need to be prototyped as soon as practicable at sev- eral locations.