Julfa District

Next 7 daysLow0%

Fire season: about 17 weeks left

Annual mean: 0detections
Peak period: mid July – mid September
Record year: 200312
Trend: −23%/ 10not significant · on raw data: +56%

Fire clock

Ring = year, slice = week. Tap a week to see its fires on the map.

Total detections per year: ODLAR's corrected record (on the VIIRS Suomi NPP scale) and the raw record (MODIS and VIIRS joined as they are)
Year	Harmonized	Raw
2001	2	0
2002	2	0
2003	12	4
2004	7	2
2005	2	0
2006	7	2
2007	10	3
2008	2	0
2009	2	0
2010	2	0
2011	2	0
2012	0	0
2013	3	3
2014	3	3
2015	1	1
2016	9	9
2017	1	1
2018	2	2
2019	4	11
2020	7	9
2021	0	2
2022	0	2
2023	1	5
2024	3	9
2025	2	4
2026	1	2

Tap a week to see the detections behind it.

Lighter: less certain (mostly the MODIS years, before 2012)No observationSatellite outage

A detection is a heat signal seen from space, not burned area or a count of separate fires.

November vs a usual year

Below normal

Weeks of 26 Oct – 8 Nov, compared with the same weeks of every year since 2001

ODLAR

0.00×

usual = 1×

This month: 0.0
Usual (median): 0.1
Rank: 16 of 26

Next 7 days

Low

Full risk view

Low

Expected detections: ~0.0

Outside the fire season↓

Sun14°

Mon9°

Tue9°

Wed11°

Thu12°

Fri12°

Sat14°

Right now

early November

This week0.0

This week is more active than 28% of past years at this time

Year to date1

#20 of 26 yearsNormal: 2

What's ahead

From the next days to a year out. The further ahead, the less certain, and the chart says which part is a forecast and which is just the usual year.

Next 7 days

Low

~0.0 expected · usual <0.1

Next 3 months

About usual

~0.0 expected · usual 0.1

likely 0.0–0.6 · usual 0.0–0.3

Season

Ends ~8 March
Peak ~16 August

ForecastLikely rangeUsualUsual rangedetections / week

How sure is this?

Daily fire weather: the Open-Meteo weather forecast, most reliable for the first 3 days.
Next 7 days: tested on 6 past years, the model ranked a week that turned busy above a quiet one 86 times in 100 (the season alone: 72).
Next 3 months: the real total fell inside the likely range 76 times in 100 (the usual range alone: 71).
Beyond 3 months: the usual year. 8 in 10 past years fell inside the grey band. Weather that far ahead is not known.

Season

mid July – early March

Onset: mid July
Peak: mid August
End: early March

Two-peak season: spring and summer–autumn

Critical windows

mid July – mid September
52% of annual activity
early March – mid March
17% of annual activity

The normal year vs 2026

Bands show the 25-year range; the white line is this year so far.

Compare with

25-year history

The raw record jumps in 2012 when VIIRS arrives; the harmonized record does not.

Harmonized: −23%/10Raw: +56%/10

HarmonizedNaive merged record90% uncertainty

Jump test (Pettitt): the raw record jumps in 2013 to 5.1× its earlier level (p 0.02). ODLAR's record has none (p 0.24).

Unusual years

12

2003
12
6.1× the normal annual activity
peak 11× stronger than normal
season ended 25 weeks early
2004
6
peak 4.7× stronger than normal
season began 5 weeks late
2006
7
peak 5.5× stronger than normal
season ended 6 weeks early
2007
10
5.0× the normal annual activity
peak 8.4× stronger than normal
season began 3 weeks early
season ended 20 weeks early
2013
3
season began 3 weeks late
season ended 18 weeks early
2015
1
season began 28 weeks late
2017
1
season began 32 weeks late
2018
2
season began 29 weeks late
2020
7
peak 5.0× stronger than normal
2023
1
season began 23 weeks late
season ended 7 weeks early
2024
3
season began 9 weeks early
season ended 19 weeks early
2026
1
season began 28 weeks late

ReportExport dataweekly.csv annual.csv hotspots.geojson

Julfa District

Next 7 daysLow0%

Fire season: about 17 weeks left

Annual mean: 0detections
Peak period: mid July – mid September
Record year: 200312
Trend: −23%/ 10not significant · on raw data: +56%

Fire clock

Ring = year, slice = week. Tap a week to see its fires on the map.

Total detections per year: ODLAR's corrected record (on the VIIRS Suomi NPP scale) and the raw record (MODIS and VIIRS joined as they are)
Year	Harmonized	Raw
2001	2	0
2002	2	0
2003	12	4
2004	7	2
2005	2	0
2006	7	2
2007	10	3
2008	2	0
2009	2	0
2010	2	0
2011	2	0
2012	0	0
2013	3	3
2014	3	3
2015	1	1
2016	9	9
2017	1	1
2018	2	2
2019	4	11
2020	7	9
2021	0	2
2022	0	2
2023	1	5
2024	3	9
2025	2	4
2026	1	2

Tap a week to see the detections behind it.

Lighter: less certain (mostly the MODIS years, before 2012)No observationSatellite outage

A detection is a heat signal seen from space, not burned area or a count of separate fires.

November vs a usual year

Below normal

Weeks of 26 Oct – 8 Nov, compared with the same weeks of every year since 2001

ODLAR

0.00×

usual = 1×

This month: 0.0
Usual (median): 0.1
Rank: 16 of 26

Next 7 days

Low

Full risk view

Low

Expected detections: ~0.0

Outside the fire season↓

Sun14°

Mon9°

Tue9°

Wed11°

Thu12°

Fri12°

Sat14°

Right now

early November

This week0.0

This week is more active than 28% of past years at this time

Year to date1

#20 of 26 yearsNormal: 2

What's ahead

From the next days to a year out. The further ahead, the less certain, and the chart says which part is a forecast and which is just the usual year.

Next 7 days

Low

~0.0 expected · usual <0.1

Next 3 months

About usual

~0.0 expected · usual 0.1

likely 0.0–0.6 · usual 0.0–0.3

Season

Ends ~8 March
Peak ~16 August

ForecastLikely rangeUsualUsual rangedetections / week

How sure is this?

Daily fire weather: the Open-Meteo weather forecast, most reliable for the first 3 days.
Next 7 days: tested on 6 past years, the model ranked a week that turned busy above a quiet one 86 times in 100 (the season alone: 72).
Next 3 months: the real total fell inside the likely range 76 times in 100 (the usual range alone: 71).
Beyond 3 months: the usual year. 8 in 10 past years fell inside the grey band. Weather that far ahead is not known.

Season

mid July – early March

Onset: mid July
Peak: mid August
End: early March

Two-peak season: spring and summer–autumn

Critical windows

mid July – mid September
52% of annual activity
early March – mid March
17% of annual activity

The normal year vs 2026

Bands show the 25-year range; the white line is this year so far.

Compare with

25-year history

The raw record jumps in 2012 when VIIRS arrives; the harmonized record does not.

Harmonized: −23%/10Raw: +56%/10

HarmonizedNaive merged record90% uncertainty

Jump test (Pettitt): the raw record jumps in 2013 to 5.1× its earlier level (p 0.02). ODLAR's record has none (p 0.24).

Unusual years

12

2003
12
6.1× the normal annual activity
peak 11× stronger than normal
season ended 25 weeks early
2004
6
peak 4.7× stronger than normal
season began 5 weeks late
2006
7
peak 5.5× stronger than normal
season ended 6 weeks early
2007
10
5.0× the normal annual activity
peak 8.4× stronger than normal
season began 3 weeks early
season ended 20 weeks early
2013
3
season began 3 weeks late
season ended 18 weeks early
2015
1
season began 28 weeks late
2017
1
season began 32 weeks late
2018
2
season began 29 weeks late
2020
7
peak 5.0× stronger than normal
2023
1
season began 23 weeks late
season ended 7 weeks early
2024
3
season began 9 weeks early
season ended 19 weeks early
2026
1
season began 28 weeks late

ReportExport dataweekly.csv annual.csv hotspots.geojson