Areal Evaluation of Agroclimatic Characteristics of the Hokuriku District

Abstract

This paper intends to obtain the detailed map of number of days with snowcover based on mesh-square technique and to apply it for land use planning for barley cultivation in the Hokuriku District.Statistical relationships were established between the snowcover properties and topographical land features for selected locations in the Hokuriku District. The third degree orographic mesh data was used to determine the topographical parameters which, in tern, were related to number of days with snowcover. The detailed geographical distribution of number of days with snowcover was then estimated from these relationships with the orographic mesh data as inputs. The results were applied for land use planning for barley cultivation.Important results obtained were as follows:1) Number of days with snowcover (Sd) on average was given bySd=62+2.153√Hm+0.032(Re, r1)2+0.145K+0.270(Lr15)-1.053√P60, r15-0.007Ha, r10-0.013Ks, r3 (r=0.96)Sd on five- and twenty-year return periods were respectivelySd, 5=72+2.166√Hm+0.028(Re, r1)2+0.150K+0.295(Lr15)-2.757ln(P60, r15)-0.008Ha, r10-0.016Ks, r3 (r=0.95)Sd, 20=73+6.850ln(Hm)+0.370(Re, r1)2+0.136K+0.218(Lr15)-0.013Ks, r1-0.015Ks, r3 (r=0.94)where each predictor variable is indicated in Table 1.2) Errors of estimate with above regression equations were within about 10% of the observed values for most sample stations (Fig. 3). The regressions underestimated by more than 10% for the adjacent regions of a mountain range over the elevation of about 2, 000 meters in the central Japan where especially heavy snowfall occurs. On the other hand, overestimates were obtained for those stations in island and peninsula regions where oceanic influence on climate may exist. Such deficiency suggests the necessity of introducing factors based on a larger geographical scale if more accurate estimates are required.3) Multiple regression equations (Eq. 2 through 4) provided distribution of number of days with snowcover (a part of which is mapped in Fig. 4) for all the mesh points of about 1km2 grid for the Hokuriku District.4) This information, together with the effect of snowcover on barley yield (Fig. 5), revealed the areas suitable for barley cultivation on much smaller spatial scale than before (Fig. 6).We therefore conclude that the multiple regression analysis used in this study with topographical parameters as inputs is a useful tool for agricultural land use planning.