Volume based expression of soil variables: Good descriptor of soil-plant interrelationships

摘要

This research compared the gravimetric method of expression of soil nutrient concentration with actual soil volume based expression in terms of yield response using rubber as a test crop. A rubber plantation (planted in year 1989 under a NPK fertilizer trial) was investigated. The experiment was laid out with thirteen treatments in three replications in a randomized block design planted with rubber (Clone RRII 105) at a spacing of 4.9 × 4.9 m. Twenty-four trees formed the main plot and 8 trees were used for measurements of plant related and soil parameters. The treatments included selected combinations of N, P and K; three rates of N (30, 60 and 90 kg ha-1year-1), two rates of P (30 and 60 kg ha-1year-1) and two rates of elemental potassium (20 and 40 kg ha-1year-1) and a control (without fertilizer application). Urea, rock phosphate and muriate of potash were applied as source of elements, N, P and K, respectively. All the nutrients were supplied in two equal splits during April-May and September-October every year. During August 1999, soil samples were collected from all the 39 plots from a depth of 0 to 30 cm. From all these plots rubber latex yield data were also recorded simultaneously at the time of soil sampling. Results revealed that soils vary in plant exploitable actual soil volume (ASV) and its contents. However soil nutrients are not widely expressed on volume basis and gravimetric expression can be misleading when ASV is not known. Soils had different descriptive statistics of variables when expressed in both ways. No gravimetric variables were correlated with rubber yield while volumetric Ca correlated significantly (r = 0.576**). Path analysis showed that log P, log Ca, Mg, log Mn and log Fe had prominent direct effects (0.521, 0.944, 0.454, -0.595 and -0.735, respectively) and variable indirect effects through other variables. However, log Ca only yielded significant correlation with yield even after partial neutralization of direct effect with indirect effects. Path analysis revealed prominent interactions through indirect effects although some correlations (sum of direct and indirect effects) were insignificant and described 57% variability indicating the scope for inclusion of other variables to describe yield. The study provided a lead indicating the merit of volumetric expression over gravimetric expression to explore further as these observations are of practical significance in soils with spatially varied contents of coarse fragments.