Improve Understanding of the Variation of the Current Climate and its Impact on Agriculture

2.1. COLLECT AND ANALYSE INFORMATION ON ECOSYSTEMS

To study the impact of climate change on agriculture and forestry and improve our understanding of certain climate mechanisms, it is important to gather numerous data on ecosystems (inventories of land use per species, phenological observations or production statistics) to evaluate ecosystem trends in recent decades at the landscape scales. The remote sensing is now a valuable tool in obtaining spatialised information on areas of the planet where ground measurements are difficult. Moreover, additional spatial information is essential in establishing sensitivity to water excess or deficit, water and wind erosion, and the risks of salinization.

2.2. STUDY THE LONG CHRONOLOGICAL SERIES OF CLIMATIC AND PHENOLOGICAL DATA

The national meteorological services (NMS) must adopt an ongoing policy of research and documentation of existing historical data which enrich national climate heritage. Indeed, it is important to study the behaviour of extended series of meteorological measurements on national territories over a period which extends from the end of the nineteenth century to the present day. To detect long term trends, the raw data must be processed using statistical methods designed to constitute homogeneous chronological series, which mean isolating the climatic signal from the other signals linked to modifications of observation techniques, relocations of measurement areas, or modifications of their environment (Moisselin et al., 2001) The analysis of these so-called "homogenised" series on a particular territory would then allow us to determine the significant variations of each parameter which may be relevant for the type of crop studied on the annual, seasonal or monthly scale and highlight spatial particularities. This kind of analysis generally shows that there is no single trend observed on the national scale (Moisselin et al., 2002). This type of study is essential when, for example, one considers the behaviour of crops in response to temperature, to its cumulative or limiting effects (frost or high temperature). Figure 1 shows the annual number of days with frost and heat waves observed in the period 1901-2000 at the meteorological station in Marseille (France). The regular decline of the number of days of frost between the beginning and end of the twentieth century is quite spectacular. The increase of the number of days of heat wave is more marked over the last two decades, which

(a)

Figure 1. (a) and (b) Annual number of days with daily minimal temperature (Tn) below 0 °C and daily maximal temperature (Tx) above 25 °C over the period 1901-2000 at the weather station in Marseille (France).

Figure 1. (a) and (b) Annual number of days with daily minimal temperature (Tn) below 0 °C and daily maximal temperature (Tx) above 25 °C over the period 1901-2000 at the weather station in Marseille (France).

is consistent with the results in the report made in 2001 by the Intergovernmental Panel of experts on Climate Change (IPCC) as regarding to average temperature (Houghton et al., 2001). The processing of extended series of climatic data should also allow a validation of the regional climate simulation models.

These extended data series should also to be complemented by phenological series coming either from observations of the natural vegetation or forest species, or from cultivated species and particularly for perennial ones (fruit trees or vines). The analysis of the latter is, of course, of great importance in supporting the phenological models which constitute the basic modules for crop simulation models (as we

29/04 24/04 19/04 14/04 09/04 04/04 30/03 25/03 20/03 15/03

*

*

*

*

* '

———-♦

* *

R! = 0,1763

1973 1976 1979 19B2 1985 19B8 1991 1994 1997 2000

1973 1976 1979 19B2 1985 19B8 1991 1994 1997 2000

Figure 2. Flowering dates (stage F1) of the apple tree in south-east France (Balandran) over the period 1974-2001 (according to Domergue, 2001).

R2 = 0,4919

1939 1945 1951 1957 1963 1969 1975 1981 1987 1993 1999 Figure 3. Dates of the beginning of the grape harvest in Chateauneuf du Pape (South of France) (according to Ganichot, 2002).

1939 1945 1951 1957 1963 1969 1975 1981 1987 1993 1999 Figure 3. Dates of the beginning of the grape harvest in Chateauneuf du Pape (South of France) (according to Ganichot, 2002).

shall discuss later). Thus, by way of an example, Figure 2 shows the evolution of the flowering dates of the apple tree in south east France (Balandran) over the period 1974-2001 (Domergue, 2001) and Figure 3 that of the grape harvest dates in Chateauneuf du Pape in the south of France from 1940 to 2000 (Ganichot, 2002). These two examples show an evolution in the phenological precocity of the various crop stages but also large interannual variability. It is therefore appropriate to be careful in analysing these results as to the partitioning of causes between growing techniques and climate interannual variability.

0 0

Post a comment