Use of Magnetic-Susceptibility Measurements in Assessing Soil Uniformity in Chronosequence Studies Pinchas Fine,* Michael J. Singer, and Kenneth L. Verosub ABSTRACT Recent observations show that soils with vertical morphological dif- ferentiation often exhibit a distinct ferrimagnetic profile with the mag- netic susceptibility (MS) of eluvial horizons enhanced with respect to that of illuvial, Cr, or R horizons. In addition, pedogenic ferrimag- netic material is mostly superparamagnetic (SP) and single domain (SD) (<30 nm), whereas inherited titanomagnetites are predominantly multidomain (MD) grains. We suggest that the continuous pedogenic development of a soil profile in homogeneous parent material results in enhancement of MS and in a gradual shift from MD to SD and SP magnetic states, and that deviations from this pattern arise from dis- continuities in the soil or from climatic or topographic changes. This hypothesis was tested on 18 northern California pedons, formed in four marine and river terrace systems and in volcanic parent material. The ages of the terraces range from 600 yr to 1.6 million yr and the MS of the parent materials range from 10 to >1000 x 10-" m 3 kg' 1 . Established chronosequence members were used to determine that enhancement of MS and shifts in magnetic domain state are correlated with age. Exceptionally large or small magnetic susceptibilities throughout the soil profile, lack of magnetic enhancement, and dis- continuous magnetic profiles in other soils could each be related to morphologically or chemically determined discontinuities. Thus, mag- netic susceptibility, an easily measured soil attribute, can be used to evaluate the suitability of pedons for chronosequence studies. A SSESSMENT of soil and parent material uniformity zV is important in the study of postincisive chron- osequences (Birkeland, 1984). Only when inter- and intra-profile uniformity is established can variations in parameters of profile development be attributed to soil age and soil-forming processes. In addition to parent material uniformity within a chronosequence, site sta- bility and uniformity of soil-forming processes are re- quired to isolate time as a soil-forming factor. Few chronosequences meet all of these requirements, and some environmental changes are inevitable or tolera- ble as long as the rates of change are smaller than those of soil formation. There is a need for a quick method to assess the suitability of a soil for chron- osequence studies. One method is to look carefully for depositional discontinuities in the soil profile. Un- fortunately, in many cases finer lithological or envi- ronmental discontinuities are revealed only after tedious chemical or physical analyses are done, or they are inferred from a lack of consistency in laboratory re- sults. Various parameters can be used to test for uniform- ity (Barshad, 1964), the more common of which are the ratios of resistant minerals and associated elements in the nonclay fractions (Mitchell, 1975) and the par- P. Fine, Inst. of Soils and Water, ARO, The Volcani Center, Bet Dagan 50250, Israel; and M.J. Singer, Dep. of Land, Air, and Water Resources, and K.L. Verosub, Dep. of Geology, Univ. of California, Davis, CA 95616. Joint contribution from the Dep. of Land, Air, and Water Resources, Univ. of California, Davis and the Inst. of Soils and Water, the Volcani Center, Bet Dagan. Received 14 Dec. 1990. 'Corresponding author. Published in Soil Sci. Soc. Am. J. 56:1195-1199 (1992). ticle-size distributions of silt and sand fractions (Raad and Protz, 1971; Busacca and Singer, 1989). We have previously shown that continuous soil de- velopment may result in a predictable pattern of MS (or x) distribution in the soil profile. The MS of elu- vial horizons is enhanced with respect to that of illu- vial, Cr, and R horizons, the extent of the MS enhancement being indicative of soil-forming processes (Fine et al., 1989; Singer and Fine, 1989). Newly formed pedogenic (or authigenic) ferrimagnetic min- erals are found predominantly in SP (<25 nm) to SD (<70 nm) magnetic grain sizes (Mullins, 1977), al- though they can occur associated with all soil particle- size separates (Fine et al., 1989). The pedogenic ferrimagnetic minerals gradually re- place the inherited, sand-sized titanomagnetites (Thompson and Oldfield, 1986; Maher, 1986). One easily measured attribute of ferrimagnetic particles at the SP-SD boundary (=15-30 nm) is the frequency dependence of the MS (XFD)- This is the percentage of loss of MS following a 10-fold increase in the fre- quency of the external inducing magnetic field. The XFD of SP and small SD grains is large, whereas that of coarser grains is small or zero. Another feature of the ferrimagnetic material that is a secondary Fe ox- yhydroxide is its solubility in the CBD treatment. Hence, the fraction of soil MS that is removed by CBD treatment is attributed to the pedogenic ferri- magnetic component (Fine and Singer, 1989). It should be mentioned that the larger XFD values and the larger proportion of MS removal by the CBD treatment in the eluvial horizons than in their illuvial counterparts are the consequence of contradicting processes. One is that the eluvial zone is the favored site for the authigenic formation of the ferrimagnetic material (Mullins, 1977). In addition, the weathering of the CBD-resistant, inherited titanomagnetites is more intense in the elluvial horizons (Fine et al., 1989). The contradicting process is the possible illuviation of the fine-textured authigenic ferrimagnetic material. Yet, the illuviation is often retarded by agglomeration of this ferrimagnetic material into miniconcretions (Fine et al., 1989). As is demonstrated below, the magnetic parameters MS and XFD are sufficient in delineating soil uniform- ity and continuity in chronosequence studies. The loss of MS in the CBD treatment can be considered equiv- alent to the XFD parameter. Characteristic MS en- hancement in any one pedon or within members of a chronosequence is expected to coincide with a shift in the relative contributions to MS from inherited (geo- logic) vs. pedogenic components. Deviations from the expected trend are indicative of a discontinuity in the parent material, erosion of the soil surface, deposition of imported material, or persistence of waterlogging. Abbreviations: MS, mass magnetic susceptibility; SP, superpar- amagnetic; SD, single domain; MD, multidomain; CBD, citrate- bicarbonate-dithionite; XFD. frequency dependence of the mag- netic susceptibility. 1195