A bio-geophysical model of Net Primary Productivity from the climatologic chart concept

A bio-geophysical model of Net Primary Productivity has been presented for the first time at the 2nd Fire behaviour and fuel conference at Dustin Florida (March 2007) Reference: Wildfire, weather and productivity (M.L. Bernard, N. Nimour) In: Butler, Bret W.; Cook, Wayne, comps. 2007. The fire environment--innovations, management, and policy: conference proceedings. 26-30 March 2007; Destin, FL. Proceedings RMRS-P-XX. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. XXX p. CD-ROM and DVD.

In the past, Lieth and Whittaker (1975) summarized the net primary productivity of 4 large zones in the world, each of them containing from 12 to 20 ecoregions with their climatic data of precipitation in mm and temperatures in Celsius degrees.

For each zone, we have analyzed the correlation between net primary productivity with precipitation and with Kelvin temperature according to the Arrhenius term (Table 2). Identifying the two correlations, we obtain an expression of precipitation as function of 1 / T (Table 3). For the three first zones, the agreement is satisfactory in spite of their low square regression coefficients.

We have particularly examined the 4th zone results, which has the best square regression coefficient. In this zone, the net primary productivity is among the highest of the world, up to four thousand of grams per square meters by year (Lieth 1962).

So the net primary productivity of zone 4 as function of temperature is:

NPP2 – NPP1 = 6* 106 (1 / T2 -1 / T1)

P expressed in g/ m2/year and T in Kelvin degrees.

In the same publication, Lieth and whittaker (1975) proposed an empirical equation of productivity Y as a function of Celsius degrees X:

Y = 3,000 / ( 1 + e1.315 – 0.119X )

Converting temperature data from Celsius to Kelvin degrees, we obtain an expression of productivity containing an Arrhenius term:

NPP2 – NPP1 = Constant (1 / T2 -1 / T1)

Taking 10 values of temperature between -13 °c and +28 °c, we obtain an Arrhenius law of net primary productivity with a square correlation coefficient very close to 1 (Figure 11).

NPP = 5*106 /T + 19,320 (R2 = 0.979)

Concerning the Wildfire, Net Primary Productivity and N.D.V.I. :

The values of the normalized difference vegetation index (NDVI) extracted from satellite sensor data acquired by the National Oceanic and Atmospheric Administration -Advance Very High Resolution Radiometer (NOAA – AVHRR) have often been used for estimating forest fire risks. According to Fang 2001, to examine the relationship between variability of NPP and precipitation at a broad scale, a long term NDVI data set coupled with a historic climate data set constitute a useful and powerful data source, because NDVI data are strongly correlated with terrestrial NPP.

The relative greenness (R.G.) was described by Burgan and Hartford (1993) as expressing how green each cell is relative to the range of greenness. Relative greenness which has been used by Burgan and others (1998) for the evaluation of fire potential index (F.P.I.), indicates that it is a potentially valuable fire management tool for land management agencies.

Gabban (2006) proposed a use of a new index referring to the dynamic relative greenness index (DRGI), to consider the inter-annual variability of NDVI at each precise location within the studied region. It was computed over the European Union countries and tested in Spain, France, and Italy. DRGI values and the number of fires were fitted using an exponential model. The testing of the DRGI over the 13 year period showed a very high correlation (R2 = 0.94) between the number of fire outbreaks and the level of risk determined by the index. So, this statement introduces the number of fires on a logarithmic form as a linear increasing function of DRGI

As a conclusion for this poster:

It was established that the fire monthly frequencies are linearly correlated with precipitation. With regard to the temperature, the fires frequencies are narrowly correlated with the Arrhenius factor (1/T, T in Kelvin degrees) translating the existence of a regulating photochemical activation phenomenon (PAR or UVB according to the season). All these correlations can be found on a multi-annual scale.

The reconstruction of the studied area climate can be carried out by identifying the expressions of the frequency of fires as functions of precipitation and temperature. We proposed the concept of “climatologic chart” (Bernard 2004) starting from such reconstruction process. This graphical representation method of the “climatologic chart of a region, that we introduced starting from the climate data, P= F (1/T) is fully justified by the highlighted vegetation production laws, independently of any knowledge concerning the existence of wildfires in the considered region.

Indeed, the Net Primary Productivity is expressed in function of precipitation or temperature, through laws similar to those found upon the fires frequency in log scale on an inter-annual scale. That brings an answer to the questions that we had on the role of the biomass production in the fires frequency laws. Moreover, the N.D.V.I and its inter-annual variability, which are closely linked to the vegetation production and the fires statistics, still reinforce this similarity.

The proposed model set will constitute valuable tools for the study of the climatic changes and their repercussions on the fires regimes, aridity, and stability of vegetation communities and may contribute to develop new fire management strategies as exposed in the San Diego declaration on climate changes and fire management.