Light Curve

Astronomers decide whether a star is truly dimming by asking if the dimming could be luck. Do the same for Earth: pick a signal, watch its light curve, then test whether the trend is real or just noise.

Loading the verdict…

Where should you point the telescope?

Pick a signal

Load real NASA data (a Giovanni or AppEEARS CSV)

Fastest real-data start: open NASA GISS’s GISTEMP global-mean temperature CSV, select all, copy, and paste it below. The year-by-month table is detected automatically, and the window starts at the latest 25 years (drag From to go further back).

Paste monthly values as date,value lines (for example 2010-03,4.2), or load a CSV file.

Drag any slider. The chart, the tests and the galaxy respond instantly: stronger signals spin the galaxy faster and warm its light, and your pointer pushes the stars aside.

Bring real NASA data in five steps

  1. Open Giovanni and choose Time Series, Area-Averaged.
  2. Set a date range of ten years or more, and draw a box around your region.
  3. Pick a monthly variable, for example precipitation (GPM), aerosol optical depth (MODIS), or soil moisture (GLDAS).
  4. Plot the data, then use Download Data and choose CSV.
  5. Paste or load the CSV in the panel above and press Analyze my data. Fill values such as −9999 are skipped automatically.

Menu names on Giovanni can change, so follow its on-screen help if a label differs.

Reading the results

Plain-language guide to each number

Keep investigating

About this project

What it does. Light Curve helps students and curious people act as data detectives: find a variable, load its monthly time series, see the trend, and test whether it is real. It was made for the NASA Space Apps Challenge on detecting trends in Earth science data.

NASA data and resources. The tool is built around time series exported from NASA’s Giovanni (GES DISC) and the global temperature tables from NASA GISS (GISTEMP v4; credit: NASA GISS/GISTEMP). The first built-in demo is the real GISTEMP v4 global-mean monthly table, January 1980 to October 2026 (GISTEMP Team, 2026, NASA Goddard Institute for Space Studies, dataset accessed 2026-11-14 at https://data.giss.nasa.gov/gistemp/). The variable finder points to MODIS, OMI, AIRS, GPM IMERG, MERRA-2 and GLDAS products, and to Hydrology Time Series and AppEEARS. Partner and outside references: CSA (SWOT mission page, Geo.ca), KNMI Climate Explorer, NOAA Climate Data Online.

Tools. Plain HTML, CSS, JavaScript, SVG and Canvas with no outside libraries. The statistics (least-squares fit with t-test and confidence interval, Mann-Kendall with Sen’s slope, and a 2,000-shuffle permutation test) are written in the page itself.

AI disclosure. The theme and concept were chosen by the team. The code, interface text and this disclosure were written with AI assistance (Claude, by Anthropic) and reviewed by the team. The first demo uses real NASA GISTEMP data; the other three demo series are generated by code, are labeled as simulated, and are not NASA measurements. The starfield and galaxy are drawn by code. This page contains no AI-generated images, audio or video, and no NASA logos or branding.

Limits. It tests straight-line trends only. Monthly data are often correlated from month to month, which can make tests look more certain than they are. Mann-Kendall is used without a correction for tied values. Treat results as a starting point for investigation.

Except for the first demo (real NASA GISTEMP data), the demo series are simulated to illustrate typical behavior (a seasonal cycle, noise, and a chosen trend). They are not NASA measurements. For real results, download a series from Giovanni and load it above. NASA does not endorse non-U.S. Government websites and is not responsible for their content; follow each site’s data use terms.