A few number of soil profile studies have pointed out that Joso Clay, a thick clayey layer underlying the shallow groundwater bodies in the diluvial uplands of southern Ibaraki and northern Chiba prefectures, Japan, has a structured surface layer in which the hydraulic conductivity is one to several orders of magnitude higher than that of the underlying massive layer. The spatial distribution of the boundary between the structured and massive clay layers should be determined as the aquifer-aquitard boundary; however, such attempts have not been made. In the present study, dynamic cone penetrometry was used to determine the spatial distribution of the aquifer-aquitard boundary in Joso Clay at a field scale in Tsukuba-Inashiki Upland, Ibaraki prefecture, Japan. The dynamic cone penetration index N d5 , every 5-cm depth average of the number of hammer strikes required for the tip cone to enter into the soil per unit of depth, was determined to a depth of 3.1-3.7 m at 54 rectangular grid points with approximately 5-m spacing in a field with an area of 943 m 2 . The field-scale N d5 profile data suggested that prismatic structures with N d5 values of up to 50-60 m -1 are widely distributed in the uppermost layer of the Joso Clay. Accordingly, a criterion of N d5 > 60 m -1 was proposed for determining the underlying aquitard in this field. The depth to the aquifer-aquitard boundary was determined to be 2.58 m on average, ranging from 2.40 to 2.89 m. Furthermore, annual water table monitoring at each grid point showed that the thickness of the structured layer, varying from 0.30 to 0.75 m, occupied 61-100% of the annual average shallow aquifer thickness, which varied from 0.40 to 0.87 m. These results demonstrate the significance of the uppermost structured layer in the Joso Clay as the major part of the shallow aquifer at a field scale in this upland.

The effects of wellbore-wall compression from rough excavation on monitored groundwater levels and qualities under natural hydraulic gradient conditions were investigated in a shallow clayey Andisol aquifer. Nine wellbores reaching the underlying aquitard at about 2.6-m depth were constructed by dynamic cone penetrometry to mimic rough wellbore construction. Five of these were constructed under wet aquifer soil conditions and the remaining four under dry conditions. A 15-month period monitoring showed that the groundwater levels in the wellbores constructed under wet conditions responded significantly in retard of, and in narrower ranges than, those constructed under dry conditions. The wellbore-wall hydraulic conductivities at the former wellbores were calculated to be more than one to two orders of magnitude lower than those at the latter ones. Furthermore, remarkable nitrate removal attributable to the occurrence of a heterotrophic denitrification was observed in one of the former wellbores. In contrast, the groundwater levels and qualities in the latter wellbores appeared to be generally similar to those monitored in the conventional soil coring and augering-derived wellbores. Our results suggest that the wellbore-wall compression induced by rough excavation under wet and soft aquifer soil conditions leads to a substantial decrease in the wellbore-wall hydraulic conductivity, which in turn can lead to unreliable groundwater levels and qualities. This problem can occur in clayey Andisols whenever the aquifer soil is wet; however, the problem can be largely avoided by constructing the wellbore under dry and hard aquifer soil conditions.