WHEAT GROWING

Timaru Herald, Volume CXXXVII, Issue 19329, 2 November 1932, Page 9

WHEAT GROWING

V. SOIL MOISTURE AND WHEAT YIELDS. ((By the Wheat Research Institute.) Every farmer knows that the yield of wheat depends on the soil moisture, and that soil moisture depends on rainfall at the proper time. Unfortunately our weather records do not distinguish between rain at the proper time, and at other times. Annual rainfalls have little relation to wheat yields as the following figures will show. They represent average wheat yields at Lincoln College and the total rainfall in the preceding year.

Here we see that in 1922, after a dry year, we had 11 bushels per acre more than after the preceding wet year; and in 1926 the yield was practically the same as for the preceding year, although there were 10 inches more rain. Seasonal Rains. Records have been kept at Lincoln College which enable us to correlate the wheat yield with the rainfall during the growing season, and also to separate the effect of winter rainfall from that of spring rainfall and so on. These calculations, and many others, were made in 1929 by Dr. Kidson, director of the meteorological office, using 27 years of records of rainfall and Wheat yield at Lincoln College. When series of figures that may express cause and effect are compared a final figure called the coefficient of correlation is obtained. If this figure is plus 1 it means that all the efiect was produced by the given cause; if it is plus 5 It means that half the effect was produced by that cause, and so on. If, on tne other hand, there Is a minus sign before the figure it means that the effect was hindered instead of helped by the cause considered. Thus if the correlation between wheat yield and winter rain is minus 2 it means that the more winter rain the less the wheat yield, and that the winter rain was responsible for about onefifth of the loss of crop. Now taking Dr. Kidson’s figures, we find that the correlation between wheat yield at Lincoln and the rainfall there from April to September was minus 17. That is the rainfall up to September has very little inuence on the yield, but that to some slight extent heavy winter rains decrease yield, probably because over-wet soil hinders root development. On the other hand, the correlation between yield and rainfall from October to January was plus 6. that is heavy spring and summer rains increased the yields, and were responsible for over half of the increase above the normal. From these figures we can deduce, if we did not know it perfectly well already, that a moderately dry winter and a moist spring and summer give us the highest wheat yields. Usefulness of Rain. So far so good. But the usefulness of rain in spring is influenced by the amount of rain there has been in the winter. This is made very clear by the records from Timaru during 193132.

Here it is seen that there was a drought in winter. Then the spring came with normal rainfall, but this was all used up as it fell. Then the summer drought came and there were no reserves of moisture, and the crops perished. It thus seems that rainfall even when divided into periods does not give as much information as it should as to its effect on the wheat crop. The fact is, of course, that rain may either soak into the ground or run off the surface, and even that which soaks into the ground may either remain there for the use of the crop, or be evaporated by nor’-westers. Evaporation. In an attempt to find the useful rainfall Lincoln College for some years measured the monthly evaporation. A water tank was sunk to the level of the soil, and the water was kept at a constant level. While the annual rainfall averaged 26 Inches the annual 1 evaporation averaged 34 inches. The information was interesting. but it represented only the evaporation from a water surface, not from the soil, and so did not give much information about the wheat crop. The Real Test of Moisture. The only real way to find out what effect rainfall at any time will have on the crop, is to measure the percentage of water in the soil. This is a laborious task, and so has been postponed for many years. But this year observations have been started, and 20 weekly samples are being taken from different plots in a field of wheat. This field was fallowed last summer, and so during the dry weather of i September it probably contained more j moisture than most fields did. The samples are being taken right through the soil and just into the clay about ) 10 inches down. Mr M. J. Scott gives i the information that when soil of the wheat growing type contains 20 per cent, of water, plants thrive at their best, and when the soil is either wetter or drier than that, growth falls off, and that when the soil contains 25 per cent, of moisture it is saturated and any extra rain either runs off, or lies on the surface till it is evaporated. The samples to date give the following results:—

Thus we see that even on this fallowed field the moisture in the soil was being seriously depleted by October 7th, but the rain on October 10th restored it to somewhere near the optimum, and the further rain on October 17th did not check the constant tendency to drying out of the soil. The rain of October 23rd and subsequent days, however, gave as much water as the soil could hold. The winter was dry, as the rainfall for June, July, August and September totalled only 5.75 inches. The soil is now saturated, and so we apparently have hopeful conditions for a high yield next harvest. It is intended to continue these observations over several years to find the correlation between crop yield and soil moisture. Acknowledgment is made to Lincoln College for practically all the figures in this article.

Ave. Yield at Rain of preceding Year Lincoln Col. at Lincoln Harvest Year. bus. per ac. inches. 1921 .. .. .. 38 28.3 1922 . . , 21.3 1923 .. , 24.5 1924 .. . 26.2 1925 .. 21.9 1926 .. . . .. 42 31.9

Inches of Rain Months. Average 1931-2 April, May, June, July 6.6 2.6 Aug., Sept., Oct 5.3 5.2 Nov., Dec., Jan 6.6 4.5

Per Cent Inches Date 1932 Moisture Rainfall Sept. 30th 18.7 Oct. 7th 15.7 Oct. 10 th .68 Oct. 12th 18.3 Oct. 17th .32 Oct. 20th 16.9 Oct. 21st and 22nd 1.12 Oct. 23rd 23.5 Oct. 23rd and 24th 1.08 Oct. 25th 28.1