THE TREND AND VARIABILITY IN THE EXPRESSI ON OF THE WEATHER COURSE INFLUENCE ON MAIZE YIELDS

The growing season is characterized mainly by concrete weather conditi ons. The yields of crops are dependent to a great extent upon the grow ing season and so upon the course of weather in the given year. To exp ress the influence of weather (the course of total month precipitation and month average temperatures) on yields in the presented study, we gave in the framework of the pair linear regression, to each yield (gr ain maize) of the given year (from the chosen continuous 8-year period in the chosen trial variants with interrow mechanical cultivation, se ed-bed soil mellowing and water regime) the value of the linear trend and then also the value of variability, characterizing so in two ways the mentioned course of weather - of precipitation and temperatures th rough nine months (January - September) in the same year. The trend wa s evaluated by our own method (Mazdur, 1990), evaluating the variabili ty by modification (Mazur, 1992) of original stability index (Kendal, Stuart, 1977). The original stability index was also modified due to t he fact that positive and negative value appeared simultaneously in th e evaluated sets of the nine month average temperatures. This also res ults in a further determination of the limits of calculation applicati on of the modified stability index I-s1, I-s2 (I-s) as defined in Tab. I by proportion of the negative and positive values in the whole set of the values analyzed for stability (variability). These limits are i nterpreted graphically in Figs 1 and 2. The mentioned limits are conne cted with the course of deviations in the given number n of the analyz ed values. These deviation courses can be limited by two extreme cases - the maximum asymmetries (irregularities) in the deviation distribut ion: the extreme of minimum [one negative deviation - below the averag e - is compensated by n - 1 (equal in size) positive deviations - abov e the average similar to Fig. 1] and the extreme of maximum with an in verse numerical proportion in the compensation of the positive and neg ative deviations (Fig. 2). These extremes are presented in Tab. I in i ts left (a maximum of the minimum extreme) and right (a maximum of the maximum extreme) fifth part. In the central part of the table is the proportion of the analyzed negative and positive values in the symmetr ical deviation course. In the centre of the left half of the Tab. I is the minimal asymmetry of the minimum extreme, and in the centre of th e right half of the Tab. I is that of the maximum extreme, in which th e number in the absolute values (equally big) of the negative and posi tive deviations differ mutually in one: \n(Delta+) - n(Delta-)\ = \1\, where n(Delta+)greater than or less than n(Delta-). It is necessary t o add that the original calculation I-s1, I-s2 in case of the odd numb er of n is in fact simplified and thus only approximate, leading to so me decrease of the limit extreme values compared with reality. It is o nly a misrepresentation of a causal asymmetry - of and irregular distr ibution of deviations around the average because (I-s1 + I-s2): 2 alwa ys leads to the exact value of I-s. The mentioned misrepresentation of the asymmetry expression decreases with the increase of the n-number. The above-mentioned limits of the negative value proportion can also be understood contrarywise, i.e. on the side of the minority proportio n, the positive values will be against the prevailing negative values in the analyzed set. In this case, unlike the above mentioned trend qu otation, the calculation scheme of stability indexes will be as follow s: I = 1 + 2 (k - A/B), where I = (I-s1, I-s2, I-s). It has been prove d that the trend formulation of precipitation dynamics from month to m onth in the same year (Fig. 3) is more suitable for the expression of the dependance of maizeyields upon them while in the case of temperatu res (Fig. 4) it is rather a (total) variation characteristics of their developement. The result of the analysis showed indirectly the possib ilities of the decrease of dependance of the field production upon wea ther - mainly upon temperature, and that not only with the known appli cation of sprinkling irrigation but also with high intensity of interr ow cultivation and its combination with more intensive soil mellowing in the seed-bed preparation.