Skip to main content
Fire Ecology

Table 2 Fire season and interannual variability by sample unit. HD clusters selected were those with the largest sample sizes for each RL.

From: Unauthorized Firesetting as Socioecological Disturbance: A Spatiotemporal Analysis of Incendiary Wildfires in Georgia, USA, 1987–2010

Sample unit

Fire season

% of firesa

Stand. dev. %b

Ave. size (ha)c

Stand. dev. (ha)d

Adj. R2e

P valuef

Statewide

Feb–Apr

44.54 %

11.87 %

2.86

1.18

0.640

<0.001

Statewide

Nov

9.45 %

6.74 %

3.15

4.19

0.413

<0.001

RL1

Feb–Apr

46.56 %

14.42 %

3.49

2.67

0.715

<0.001

RL1

Oct–Nov

20.05 %

13.70 %

2.32

1.64

0.942

<0.001

RL2

Feb–May

53.79 %

20.93 %

1.52

1.27

0.349

0.001

RL2

July

8.78 %

13.54 %

0.86

2.18

0.312

0.003

RL3

Jan–Apr

69.04 %

15.43 %

1.77

1.38

0.429

<0.001

HD224 (RL1)

Feb–Apr

48.73 %

21.95 %

1.63

1.23

0.482

<0.001

HD224 (RL1)

Oct–Nov

22.00 %

19.37 %

1.72

3.51

0.571

<0.001

HD197 (RL2)

Mar–May

46.27 %

23.80 %

1.03

1.20

0.490

<0.001

HD197 (RL2)

July

9.70 %

21.86 %

0.75

2.20

0.052

0.146

HD70 (RL3)

Jan–Apr

70.00 %

22.91 %

2.22

3.17

0.214

0.013

  1. a % of fires = percentage of fires from the 24 yr total that occurred during the fire season.
  2. b Stand. dev. % = the standard deviation of the annual percentage of fires that occurred during the fire season over the entire 24 yr period.
  3. c Ave. size (ha) = average fire size for the indicated fire season over the entire 24 yr period.
  4. d Stand. dev. (ha) = standard deviation of the annual average fire season-fire size for the entire 24 yr period.
  5. e Adj. R2 = the adjusted R2 value for the linear regression of the annual number of fires per fire season with the annual area burned during the fire season.
  6. f P value = probability statistic that the null hypothesis (there is no linear relationship between number of fires and area burned) is true.
Back to article page