Laserfiche WebLink
<br />0;)1224 <br /> <br />VIII. 8 <br /> <br />Station networks <br /> <br />Both the water retaining capacity of' a soil, and the type and <br />density of a water-consuming vegetative cover may vary greatly f'rom <br />one spot to another. Also, ref'erence has been made to non-unif'orm <br />disperal of' the water in soil. The inhomogeneities commonly are such <br />that a reliable value of' total soil water within an area of' interest <br />requires (1) determinations of' soil water at numerous stations; and <br />(2) a station network, also a method and f'requency of' determination <br />designed thoughtf'ully according to intended use of' results. Standard <br />statistical techniques can be used to evaluate whether a given number <br />of' soil-water determinations represents the mean hydrologic condition <br />of an area, provided the magnitude of' sampling errors and the inherent <br />variability of' the particular area are known or can reasonably be <br />expressed. <br /> <br />Both time and spilce variations of' a soil's water content are here <br />of' concern. A station-network design that is suitable f'or a unif'orm f'ield <br />crOp under sprinkler irrigation, ma:y be wholly unsuitable f'or a brushfield <br />or a forest. A design adequate to appraise moisture holding capacity of <br />a basin af'ter winter rains or snowmelt, may be inadequate to appraise <br />moisture conditions following high intensity llummer showers. A design <br />that can discover time or space differences in soil-water storage ma:y <br />be incompetent to evaluate the exact amounts of' water involved. Systematic <br />network design may be "fool-proof" in a random physical situation (as in <br />a natural field), but may be strongly biased in a systematic physical <br />situation (as in a planted forest). <br /> <br />As f'ew as f'ive stations, ,randomly selected, have been found adequate <br />f'or evaluating summer water losses in a forest by using a neutron probe, <br />where swmner precipitation was very small and where the primary concern <br />was with dif'ferences in water loss between "control" areas and treatment <br />areas. Dif'ferences in the order of 0.2 inch of water in 6.9i.nehes of <br />total loss could be demonstrated. In contrast, absolute measure of water <br />loss from a single station may have been considerably in error. A fixed <br />bias of the measuring instruments may have added to this error. <br /> <br />In general, the limitations of available soil-water data in resolving <br />problems of hydrology arise from the non-homogeneous system anQ the general <br />tendency in the past to express this system in hydrostatic rather than in <br />hydrodynamic terms. As energetics of the soil-water system come to be <br />better understood and clearly expressed, then one can expect that these <br />limitations will shrink. <br /> <br />