NOAA Regional Integrated Sciences and Assessments (RISA) Program

Web Application Development for Dynamic Drought Index Mapping

Jinyoung Rhee *, Ph.D. Candidate

Greg Carbone *, Hope Mizzell **, and Ryan Boyles ***

* Department of Geography, University of South Carolina

** SC State Climatology Office, SC Department of Natural Resources

*** State Climate Office of NC, North Carolina State University

Palmetto Chapter of the American Meteorological Society12th Annual Mini-Technical Conference, 15 May 2006

Catawba-Wateree River Basin

Catawba-Wateree River Basin

Source: Duke Power (http://www.dukepower.com/community/lakes/cw/)

Exceptional DroughtExtreme DroughtSevere DroughtModerate DroughtAbnormally Dry

Objective

To provide local scale drought monitoring tool that generates indices in various time scale

Spatial scale (state, climate division, county, drought management area of SC, 2~8-digit HUC water basin)

Time scale (short-term or long-term)

Structure

Drought IndicesDrought IndicesStreamflowStreamflow

PrecipitationPrecipitation

Raw Data andRaw Data andPercentilesPercentiles

Interpolation andInterpolation andSpatial AveragingSpatial Averaging

Numerical Numerical models (Pytmodels (Pyt

hon)hon)

Database Database (MySQL)(MySQL) Graphical User InterfaceGraphical User Interface

ClientClient(JavaScript)(JavaScript)

ServerServer(PHP)(PHP)

SVG embeddedSVG embedded

User InputUser InputMap DisplayMap Display

PHP – PHP: Hypertext PreprocessorSVG – Scalable Vector Graphics

Calculating Percentiles

Total 55 years (1950 ~ 2004)

Palmer indices and SPI: all stations, all months

Streamflow and precipitation: each station, each month

Empirical Cumulative Distribution Function

0

10

20

30

40

50

60

70

80

90

100

-4 -3 -2 -1 0 1 2 3 4

Raw data: - 0.75

Percentile: 22.78

Interpolation (IDW)

6 nearest stations

Search up to 12 nearest stations for data

No data

Next nearest station

4 km X 4 km grids

Spatial Averaging

For each spatial scale (state, climate division, county, drought management area of SC, 2~8-digit HUC water basin)

Scalable Vector Graphics (SVG)

SVG defines vector-based graphics in XML (eXtensible Markup Language) format for the Web

Source: SVG Tutorial (http://www.w3schools.com/svg)

Shapefile to SVG

Conversion of geometries and properties of ArcView Shapefiles into SVG elements using Visual Basic Application module(from www.carto.net, http://pilat.free.fr/, Yahoo SVG developer Group; modified by Drs. Michael Hodgson and Edwin Chow)

Communication between MySQL database and SVG elements using PHP and JavaScript

ShapefileSVG element with an attribute ID=“0”

MySQL

JavaScript filese.g.

blend[SPI][0]=-4.29

PHP

Classify and assign color

Show value when mouseover, etc

DOMConvert

User Input

Display type Input

PDSI, PHDI, Z Index SPI in several time scal

es Precipitation and strea

mflow in several time scales

Raw data vs. percentile blend

Map User Input

Classification method Equal interval Quantile Natural break U.S. Drought Monitor-c

omparable category For any percentile blen

d For PDSI For SPI For weekly streamflow

percentile

Exceptional DroughtExtreme DroughtSevere DroughtModerate DroughtAbnormally Dry

Map Navigation Tools

Full view New map extent Pan New map center Zoom in & out Zoom slide

(JavaScript source code by Andreas Neumann; modified by Jinyoung Rhee)

Map Layers

Visibility Choropleth map Boundary Stations used

4 km x 4 km grid

No data

Map Layers

Visibility Choropleth map Boundary Stations used

Layer Aggregation

Selected features of the same layer have been aggregated Aggregation is based on the number of 4 km x 4 km grids with

data used for spatial averaging of each feature

35.835.8

45.1945.19

7.167.16

27.6827.68Click create map again

GraphCreate Graph

Aggregated Features

TableCreate Table

Comparison: spatial variabilityMay 1999

50% PDSI and 50% 1-month SPI

Climate Division 8-digit HUC

County

Exceptional DroughtExtreme DroughtSevere DroughtModerate DroughtAbnormally Dry

Comparison: spatial variabilityJuly 2002

50% PDSI and 50% PHDI

Climate Division 8-digit HUC

County

Exceptional DroughtExtreme DroughtSevere DroughtModerate DroughtAbnormally Dry

Comparison: time scale variabilityJanuary 1995 ~ December 2004

1-month SPI 3-month SPI

6-month SPI 12-month SPI

Long term, July 2002Long term, December 1965

Short term, December 1965 Short term, July 2002

Exceptional DroughtExtreme DroughtSevere DroughtModerate DroughtAbnormally Dry

Conclusion

Customized functions have been extended gradually through the user feedback procedures: flexibility

Use of open-source software and database: suitability for cost effective and tight budget projects