James Acker NASA Goddard Earth Sciences Data and Information Services Center (GES DISC)

James AckerJames AckerNASA Goddard Earth SciencesNASA Goddard Earth Sciences

Data and Information Services Center (GES DISC)Data and Information Services Center (GES DISC)

GiovanniGiovanni http://giovanni.gsfc.nasa.gov/http://giovanni.gsfc.nasa.gov/

Web-based application Developed by the NASA Goddard Earth

Sciences Data and Information Services Center (GES DISC)

Easy to use No need to learn data formats,

programming or download large amounts of data

Customized data analyses and visualizations with only a few mouse clicks…

Aerosol from MODIS and GOCART modelAerosol from MODIS and GOCART modelParticulate Matter (PM 2.5) from AIRNow Particulate Matter (PM 2.5) from AIRNow

Ozone Hole from OMIOzone Hole from OMI

Aerosol from GOCART modelAerosol from GOCART model1010-6-6 ppmv ppmv

Carbon Monoxide from AIRSCarbon Monoxide from AIRS

Water Vapor from AIRSWater Vapor from AIRS MODIS vs SeaWiFS ChlorophyllMODIS vs SeaWiFS Chlorophyll

Giovanni Giovanni InterfacesInterfaces

CloudSatCloudSat

HIRDLSHIRDLS

MLSMLS

OMIOMI

TRMMTRMM

SeaWiFSSeaWiFS

AMSR-EAMSR-E

HALOEHALOE

TOMSTOMS

ModelsModels

ParasolParasol

CALIOPCALIOP

Data Data InputsInputs

MODISMODIS

AIRSAIRS

MISRMISR

and and more…more…

Analysis Tools: Giovanni DataAnalysis Tools: Giovanni Data

Getting Started with Giovanni Getting Started with Giovanni

Select Area of Interest

Select Display (info, unit)

Select Parameters

Select Time Period

Select Plot type

Generate Generate VisualizationVisualization

Refine constraints and edit plot preferences

Outputs: Refine/ModifyOutputs: Refine/Modify

Download Data: Download Data: files and imagesfiles and images

Giovanni data download page Giovanni data download page HDF, NetCDF, ASCII and HDF, NetCDF, ASCII and KMZs KMZs (for Google Earth)(for Google Earth)

Using Giovanni Using Giovanni to observe the oceansto observe the oceans

Available Science Data Sets

Giovanni Oceans Tools and DatasetsGiovanni Oceans Tools and Datasets

Chlorophyll concentrationChlorophyll concentration Diffuse attenuation coefficient at 490 nm Diffuse attenuation coefficient at 490 nm (K490)(K490) Normalized water-leaving radiance at 555 Normalized water-leaving radiance at 555 nm (SeaWiFS) or 551 nm (MODIS)nm (SeaWiFS) or 551 nm (MODIS) Absorption coefficient of dissolved and Absorption coefficient of dissolved and detrital matter at 443 nmdetrital matter at 443 nm Particulate backscatter coefficient at 443 Particulate backscatter coefficient at 443 nmnm Sea surface temperature (MODIS)Sea surface temperature (MODIS) Assimilated chlorophyll and other output Assimilated chlorophyll and other output fields from the NASA Ocean Biogeochemical fields from the NASA Ocean Biogeochemical Model (NOBM)Model (NOBM)

Giovanni’s ocean data is from either:Giovanni’s ocean data is from either:the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), or the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), or the Moderate Resolution Imaging Spectro-radiometer the Moderate Resolution Imaging Spectro-radiometer (MODIS)(MODIS)

Giovanni output typesGiovanni output types

Area plot (lat-lon map)Area plot (lat-lon map)

AnimationsAnimations display successive area display successive area plotsplots

Hövmoller plots – ideal for Hövmoller plots – ideal for visualization of seasonal visualization of seasonal

signals signals

Time -Time -seriesseries

Time-series analysis with Time-series analysis with GiovanniGiovanni

Five good reasons to perform Five good reasons to perform time-series analyses: time-series analyses:

• Detecting changes (trends) over timeDetecting changes (trends) over time• Assigning causation to trendsAssigning causation to trends• Distinguishing between short-term Distinguishing between short-term variability and long-term trendsvariability and long-term trends• Predicting changes in the futurePredicting changes in the future• Determining consistency of observationsDetermining consistency of observations with measurable (and significant) trendswith measurable (and significant) trends

SignificantSignificant chlorophyll trends chlorophyll trends in the global oceanin the global ocean

What is necessary for What is necessary for a useful time-series analysis?a useful time-series analysis?

1.1. A sufficiently long data setA sufficiently long data set2.2. A meaningful environmental parameterA meaningful environmental parameter3.3. Consistency of measurement Consistency of measurement

methodologymethodology4.4. Exclusion of erroneous or questionable Exclusion of erroneous or questionable

datadata5.5. ““Unbiased” statistical toolsUnbiased” statistical tools

A foolish consistency is the hobgoblin of little minds, A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines.adored by little statesmen and philosophers and divines.

- Ralph Waldo Emerson- Ralph Waldo Emerson

SeaWiFS chlorophyll data is sufficientlySeaWiFS chlorophyll data is sufficientlyaccurate for valid time-series analysis,accurate for valid time-series analysis,due to rigorous and consistent QA/QC.due to rigorous and consistent QA/QC.

Basic steps Basic steps for time-series analysis with Giovanni:for time-series analysis with Giovanni:

1.1.Choose parameter, region, and time periodChoose parameter, region, and time period2.2.Generate the “raw” time-seriesGenerate the “raw” time-series3.3.Identify outliers – consider exclusion criteriaIdentify outliers – consider exclusion criteria4.4.Acquire ASCII numerical dataAcquire ASCII numerical data5.5.Export ASCII data to MS Excel (or other statisticalExport ASCII data to MS Excel (or other statisticalpackage)package)6. Calculate trend and significance6. Calculate trend and significance

Choose Parameter, Region, and Time PeriodChoose Parameter, Region, and Time Period

ParameterParameter: chlorophyll : chlorophyll aa concentration (chl concentration (chl aa))

RegionRegion: Pacific Tehuano Wind Zone: Pacific Tehuano Wind ZoneNorth: 12.0 South: 11.0North: 12.0 South: 11.0West: -99.0West: -99.0 East: -98.0 East: -98.0

Time Period: October 1997 – December 2007Time Period: October 1997 – December 2007

Region SelectionRegion Selection

Generate “raw” time-seriesGenerate “raw” time-series

Identify outliers:Identify outliers:consider exclusion criteriaconsider exclusion criteria

??

Acquire ASCII numerical dataAcquire ASCII numerical data

ASCII ASCII data data

download download iconicon

Time Time series series imageimage

Export ASCII data to MS Excel Export ASCII data to MS Excel (or other (or other statistical analysis package)statistical analysis package)

Save the ASCII data to a text fileSave the ASCII data to a text file Open the file in ExcelOpen the file in Excel Choose “Fixed Width”Choose “Fixed Width” Choose “MDY” for Choose “MDY” for Column Data FormatColumn Data Format Data will appear in two columns in Data will appear in two columns in spreadsheetspreadsheet

MS Excel will require the Analysis MS Excel will require the Analysis Toolpak to be installed; don’t Browse, Toolpak to be installed; don’t Browse, MANAGE!MANAGE!

Calculate trend and significanceCalculate trend and significance

Significance Significance FF

River-influenced coastal zones;River-influenced coastal zones;Sampling the watershed Sampling the watershed

In Acker, McMahon, Shen, Hearty, and Casey In Acker, McMahon, Shen, Hearty, and Casey (2009), we took chl (2009), we took chl aa as proxy for river effects as proxy for river effects in general, because of the known problems in general, because of the known problems with the data in coastal zones, particularly due with the data in coastal zones, particularly due to colored dissolved organic matter (CDOM). to colored dissolved organic matter (CDOM).

Trends in chl Trends in chl aa thus indicated changes in the thus indicated changes in the influence of the river, due to the effects of river influence of the river, due to the effects of river discharge to the ocean on nutrient discharge to the ocean on nutrient concentrations, CDOM export, and sediments. concentrations, CDOM export, and sediments.

Changes in the influence of the river were Changes in the influence of the river were primarily attributed to changes in the flow primarily attributed to changes in the flow regime of the river.regime of the river.

River-influenced coastal zones;River-influenced coastal zones;Sampling the watershed Sampling the watershed

The primary two exceptions (we think) are the The primary two exceptions (we think) are the Mississippi River, with reduced nutrients – i.e., Mississippi River, with reduced nutrients – i.e., agricultural management is working [ironic in light of the agricultural management is working [ironic in light of the oil spill] and oil spill] and

The Pearl River in China, which clearly shows the effects The Pearl River in China, which clearly shows the effects of increasing fertilizer use (and pollution) in a region of of increasing fertilizer use (and pollution) in a region of heavy agricultural activityheavy agricultural activity

Our analyses also clearly showed the impact of extreme Our analyses also clearly showed the impact of extreme events (floods) on time-series analysisevents (floods) on time-series analysis

The most interesting case was the Eel River in California; The most interesting case was the Eel River in California; the high flow period was essentially unchanged, but the the high flow period was essentially unchanged, but the low flow period had an increasing trend due to the low flow period had an increasing trend due to the marine environment, indicating a decreasing flow during marine environment, indicating a decreasing flow during the low flow periodthe low flow period

Links and places of Links and places of interestinterest

GES DISC: GES DISC: http://disc.sci.gsfc.nasa.gov/ Mirador: Mirador: http://mirador.gsfc.nasa.gov/Giovanni: Giovanni: http://giovanni.gsfc.nasa.gov/

@nasa_gesdisc, @nasa_giovanni@nasa_gesdisc, @nasa_giovanni

Facebook Group: Facebook Group: NASA Giovanni: Remote Sensing Data AnalysisNASA Giovanni: Remote Sensing Data Analysis

Thank you!Thank you!

Any questions?Any questions?