Firecast System

Q: What browsers does Firecast support?

Currently, Internet Explorer 8, Mozilla Firefox 22 and Chrome 28 are supported. Other browsers may work, but have not been tested for support.

Q: I did subscribe, but I am not receiving emails.

Sometimes, your e-mail client flags some messages as SPAM, and these messages are automatically filtered out of your inbox. Add us to your secure e-mail list to avoid that problem.

Active Fire

Q: Who provides the active fire data for Firecast?

The active fire data is provided by NASA's Fire Information for Resource Management System (FIRMS)

Q: How does MODIS detect fires?

A hotspot/fire is detected by MODIS using data from the middle infrared and thermal infrared bands. These detections are produced using the same algorithm as the standard MODIS MOD14 fire and thermal anomalies product. The algorithm examines each pixel of the MODIS swath, and ultimately assigns to each one of the following classes: missing data, cloud, water, non-fire, fire, or unknown. In most cases, this thermal anomaly is a fire, but sometimes it is a volcanic eruption or the flare from a gas well. We have no way of knowing which it is based on the MODIS data alone.

Q: What size fires can be detected?

Under ideal conditions the smallest flaming fire that can be routinely detected (i.e. near 100% probability of detection) is approximately 50 square meters in size. These are conditions such as when a fire is observed at (or near) nadir on a fairly homogeneous surface, no other significant fires are nearby, and the scene is free of clouds, heavy smoke and sun glint. In any given scene the minimum detectable fire size is a function of many different variables (scan angle, biome, sun position, land surface temperature, cloud cover, amount of smoke, wind direction, etc.), so the precise value will vary slightly with these conditions.

Q: What is the delay from the time the satellite passes until an alert is emailed?

The delay is approximately 1 day or less. Data are compiled for an entire 24-hour period and then processed through an automated system. The automated system runs each day, including weekend and holidays.

Q: What does each of the columns in the text file from the active fire emails mean?

Feature - area of interest (for example, district name, protected area name, etc.)
Latitude - the north or south location on the earth measured in decimal degrees
Longitude - the east or west location of the point on the earth measured in decimal degrees
Date - date the fire was detected
Time - time the satellite passed over the location, provided in UTC
Satellite - platform that observed fire, there are two MODIS instruments - Aqua and Terra. For more information on these platforms, see the information on MODIS in the first question.
Confidence - The confidence value is based on a collection of intermediate algorithm quantities used in the detection process. The range of confidence is expressed between 0-100 (the values range from 0=very low to 100=very high). The confidence field is experimental and should be used with caution; it is likely that it will vary in meaning in different parts of the world.

Q: Where can I get additional FAQ resources on MODIS fire detection products?

For more FAQ related specifically to active fire data please see the FIRMS' FAQ page.

Fire Risk

Q: Who provides the fire risk data for Firecast?

The fire risk model was developed at Conservation International in collaboration with the University of Maryland and funded by NASA. The data are generated at Conservation International.

Q: How do I interpret the fire risk index?

We define fire risk as the average moisture content of 100 and 1000 hour time classes of fuel. This translates into the average moisture content of litter on the forest floor that takes 4 to 40 days to completely dry up. Dry litter of the 1000 hour class that reaches the threshold of 15% indicates drying conditions for over 40 days which is representative of drought conditions. Averaging the 1000 hour moisture class with the litter moisture in the 100 hour class, which indicates a drying for about 4 days, adds short term fluctuations of moisture conditions when evaluating daily fire risk. Fire risk values range from 0-100% where 15% is the indicator for dry conditions susceptible to fire.

Q: What is the resolution of the fire risk index?

The fire risk index has a temporal resolution of 24 hours and a spatial resolution of 5km.

Q: What is the "days since last rainfall" drought index?

This is a simple drought index that counts the number of days since the last 24 hour rainfall greater then 5mm. This fire-risk index is commonly used in tropical environments. Twenty days or more since a rainfall event is considered drought-like conditions.

Q: What is the "Keetch-Byram drought index"?

KBDI is a water balance drought index based on soil moisture. The index is an estimate how much rainfall is required to saturate the top 203 mm (8") of soil at any time. Moisture is lost from the soil only by evaporation based on maximum daily temperatures and the soil is saturated by rainfall events. The first 5 mm of rainfall is assumed to be absorbed by the vegetation canopy. KBDI values range from 0-203 mm where 0 is a saturated soil and no risk of drought and 203 indicates extremely dry conditions. KBDI assumes the landscape if flat and there is uniform drying. The KBDI product is calculated based on equations from Keetch and Byram (1968) and Crane (1982).

Q: How do clouds affect the calculation of fire risk?

We use a distance-weighted, temporal interpolation to fill any gaps in the temperature and relative humidity data created by the presence of clouds or gaps between MODIS swaths. The TRMM rainfall product does not have gaps from clouds because it utilizes cloud temperature and height to estimate rainfall rates.

Q: Will you provide fire risk forecasts for other regions?

We are exploring model validation to see if it is feasible to provide these fire risk forecasts to new countries like Madagascar and Indonesia.

Fire Season Severity Forecasts

Q: Who provides the fire season severity forecasts for Firecast?

The fire season severity forecasts are a result of a collaboration between University of California, Irvine, NASA Goddard Space Flight Center, Columbia University, University of Maryland, and Duke University. The work is funded by the Gordon and Betty Moore Foundation through Grant GBMF3269 and the US Agency for International Development (USAID). The data can be accessed directly here.

Q: How do I interpret the radial dials for the fire season severity forecasts?

The fire season severity index ranges from 1-100. Green indicates below average predictions of fire activity whereas orange and red indication above average activity. The line between green and orange represents the long-term average of fire activity from 2001 to present. In the Firecast system we label green as LOW, orange as MEDIUM, and red as HIGH fire season severity. The number as the bottom of the dial indicates the percentile of this year's risk relative to the long term mean.

Q: Will you provide fire season severity forecasts for other departments in Bolivia and provinces in Peru?

We are working towards providing finer resolution fire season severity alerts for Peru and potentially providing other departments in Bolivia.

Q: Will you provide fire season severity forecasts for other countries?

These forecasts are already available for states in Brazil on the University of California Irvine's fire season severity forecast website. Significant modeling and validation are required to provide these forecast to new countries such as Madagascar and Indonesia. Such expansion would be dependent on future research grants.