Abstract
Profile sampling, which causes missing or overlapping horizons, has been generally used while studying loess stratigraphy and paleoclimate. Conversely, drill sampling of thick loess can provide a relatively complete and actual record of the Quaternary climate and environmental changes. Here, paleomagnetic chronological analysis and particle size and magnetic susceptibility analyses were performed to accurately reveal the loess stratigraphy characteristics of Luochuan loess tableland in Fu County, Shaanxi Province, China. The FX core revealed that the Quaternary loess layer was 167.03 m thick and consisted of 33 layers of developed loess/paleosol sequences and 37 loess/paleosol assemblages. The paleosol horizons indicated relatively warm and humid climatic period, while loess/sand reflected cold and arid climate. Since 2.6 Ma, the climate of the region has undergone 37 warm and cold cycles trending toward an increasingly dry and cold climate. During this time, 10 paleoclimate cycles occurred during 2.6–1.6 Ma, 15 climate cycles occurred during 1.6–0.8 Ma, and 12 loess–paleosol cycles have occurred from 0.8 Ma to the present day. Our study provides a basis for accurately and comprehensively interpreting the paleoclimate and paleoenvironment during loess deposition formation.
1 Introduction
The loess–paleosol sequences of the Loess Plateau in China are typical Quaternary terrestrial deposits, characterized by a rapid deposition rate, established thickness, continuity, and high resolution at least on glacial/interglacial scale. As they contain abundant paleoclimatic information, they can provide important information on the climate changes since 2.6 Ma [1]. Therefore, the loess–paleosol sequences are crucial in paleoclimatology and stratigraphy studies. The Luochuan area, located in the middle of the Loess Plateau in the Shaanxi Province, is a large loess sedimentary basin developed on the Cenozoic denudation peneplain, surrounded by mountains and uplands. This area has been continuously receiving dust since the beginning of the Quaternary, thereby forming complete loess–paleosol sequences [2]. The continuous loess layers formed under the influence of wind, and have recorded paleoclimatic information since the onset of the Quaternary period [3,4].
Since the first report of Luochuan profiles by Liu [5], several researchers have studied the Luochuan loess tableland. Liu [6] conducted a climato-stratigraphic study on the typical Luochuan loess profiles in Shaanxi, while Heller and Liu [7] established relatively reliable magnetic polarity time scale for the first time. Moreover, An et al. [8] proposed comprehensive classification units of loess stratigraphy and discussed the history of climate change based on the magnetic susceptibility of Luochuan loess. Sun and Liu [9] redivided the Wucheng loess strata in the Luochuan profile. Further, regarding chronological considerations, Lu et al. [10] selected magnetic susceptibility as a proxy indicator for changes in the East Asian summer monsoon climate, and established a time scale for the loess–paleosol sequences using the orbital tuning method. Liu et al. [11] further ascertained the specific location of the magnetic polarity boundary in the Luochuan profile through high-density paleomagnetic sampling and testing. Several studies on paleoclimate and paleoenvironment have also been conducted. For example, Wu et al. [12] investigated the geological environment of loess deposition in the Luochuan loess tableland. Han et al. [13,14] investigated the vegetation conditions, ambient humidity, and temperature during paleosol formation based on the results of carbon and oxygen isotope analysis of calcium nodules in various paleosol layers of Lishi loess in the Luochuan loess profile. Moreover, Wen et al. [15] discussed the climate evolution information in the Loess Plateau since 2.4 Ma ago by analyzing the total carbonates, oxide ratio, and periodic element evolutions in the Luochuan loess profile. Additionally, numerous researchers have utilized the distribution characteristics of paleoclimate proxies, such as magnetic susceptibility [4], particle size [16], Rb/Sr value [17], whiteness [18], free Fe/total Fe ratio [19], phytoliths [20], snails [21], and molecular fossils [22], in the Luochuan profile to reveal the evolution of East Asian monsoon climate [23], and provided information on paleoclimate, paleoenvironment, and paleovegetation.
However, the aforementioned studies have largely focused on profile sampling, which is typically suitable for horizons with proper exposure and clear lateral continuity. Drill core sampling of thick loess can provide a relatively complete and actual record of the Quaternary climate and environmental changes. In this study, we aimed to study the paleomagnetic chronology of the loess strata using drill cores obtained in Village Xiaoyuanzi, Niuwu Township, Fu County, Yan’an City, Shaanxi Province (hereinafter referred to as FX drill cores). Moreover, we measured and analyzed the magnetic susceptibility, particle size, and loss on ignition (LOI) of the drill cores. The paleomagnetic chronology approach and environmental indices were combined to conduct an in-depth investigation on the paleoenvironmental evolutions and provide accurate and comprehensive evidence of the paleoclimate and paleoenvironment relevant to the loess formation in the study area.
2 Materials and methods
2.1 Core overview
2.1.1 FX drilling location
The FX core (36°10′45.74″ N, 109°28′52.05″ E) was located in Xiaoyuanzi Village, Niuwu Township, 23 km northeast of Fu County, Yan’an City, Shaanxi Province, China, ∼48 km from the Luochuan profile (Figure 1). The area was geomorphologically classified as a loess residual tableland area, located in the southeast part of the northern Shaanxi slope belt of the Ordos Basin, near the Weibei uplift area, and at the edge of the central paleo-uplift.
Geographical location of FX core.
2.1.2 Stratigraphic description of the FX core
The total depth of the FX core was 200.42 m, with quaternary loess strata at 0–167.03 m and Neogene red clay from 167.03 m downwards. The drill cores were almost complete, with a core recovery rate of >95%. Moreover, the cores recovered the Heilu soil at the top and Neogene red clay at the bottom, which span more than the entire Quaternary period. The deposition sequence was complete and complex, which provided abundant paleoclimate evolution information and effective geological data for the environment and paleoclimate evolution in this area since the Quaternary.
The FX core revealed 33 layers of developed loess–paleosol cycles (33 loess and 33 paleosol units) from top to bottom, including the entire Quaternary (Q4, Q3, Q2, and Q1) strata and also older Neogene strata (N). The typical lithological and stratigraphic descriptions of the FX core are listed in Table 1.
Description of the loess strata in the FX core
| Name | Features | Top plate (m) | Base plate (m) | Thickness (m) |
|---|---|---|---|---|
| Heilu soil S0 | Gray-brownish, uniform texture, abundant mycelium development, and development of insect pores and root systems | 0 | 0.92 | 0.92 |
| Loess L1 | Gray-yellowish to brown-yellowish silt and sandy loam, loose, with occasional small calcium nodules | 0.92 | 10.94 | 10.02 |
| Paleosol S1 | Red-brownish loam, uniform and dense, abundant white calcareous spots with locally observed calcium nodules at certain levels | 10.94 | 13.44 | 2.5 |
| Loess L2 | Gray-yellowish to brown-yellowish silty fine sand and sandy loam, loose, with local adequate argillaceous materials | 13.44 | 21.58 | 8.14 |
| Paleosol S2 | Red-brownish loam, locally interspersed with clay masses, with locally observed calcium nodules | 21.58 | 25.77 | 4.19 |
| Loess L3 | Gray-yellowish to brown-yellowish fine silty sand and sandy loam, loose, containing local adequate argillaceous materials and calcium nodules | 25.77 | 29.44 | 3.67 |
| Paleosol S3 | Brown-yellowish to red-brownish loam, slightly dense, locally interspersed with brown-yellowish silty fine sand, with a small number of white calcareous spots, and occasionally observed calcium nodules | 29.44 | 31.72 | 2.28 |
| Loess L4 | Gray-yellowish to brown yellowish sandy loam, loose, with abundant white calcareous spots and a few calcium nodules | 31.72 | 35.72 | 4 |
| Paleosol S4 | Red-brownish to brown reddish loam, uniform and dense, occasionally showing calcium nodules | 35.72 | 38.88 | 3.16 |
| Loess L5 | Brownish yellow sub-sandy soil, uniform, slightly dense, with local layers containing calcium nodules | 38.88 | 45.12 | 6.24 |
| Paleosol S5 | Two layers of dense brown-yellowish sandy loam sandwiched by three layers of red-brownish loam, uniform | 45.12 | 50.06 | 4.94 |
| Loess L6 | Brownish yellow sub sandy soil mixed with reddish brown sub clay, uniform, slightly dense, containing a small amount of white calcareous spots, occasionally with calcium nodules | 50.06 | 58.17 | 8.11 |
| Paleosol S6 | Reddish brown loam mixed with grayish yellow loam in the middle, uniform and dense, with a small amount of white calcareous spots | 58.17 | 59.81 | 1.64 |
| Loess L7 | Grayish yellow sub sandy soil, uniform and dense, with a lot of white calcareous spots in the middle and upper part | 59.81 | 62.84 | 3.03 |
| Paleosol S7 | Reddish brown loam, uniform and dense, containing a lot of white calcareous spots, occasionally with calcium nodules | 62.84 | 64.69 | 1.85 |
| Loess L8 | Brownish yellow to grayish yellow sub sandy soil, uniform, slightly dense, containing a small amount of white calcareous spots, occasionally with calcium nodules | 64.69 | 65.94 | 1.25 |
| Paleosol S8 | Red-brownish loam, uniform and dense, with a few white calcareous spots | 65.94 | 68.31 | 2.37 |
| Loess L9 | Gray-yellowish to brown-yellowish silty fine sand, with locally distributed sandy loam, containing a small number of white calcareous spots and calcium nodules | 68.31 | 79.91 | 11.6 |
| Paleosol S9 | Red-brownish loam, uniform and dense, with a few white calcareous spots or calcium nodules | 79.91 | 81.13 | 1.22 |
| Loess L10 | Brownish yellow sub-sandy soil, uniform, slightly dense, with a certain amount of white calcareous spots | 81.13 | 83.07 | 1.94 |
| Paleosol S10 | Red-brownish loam, uniform and dense, with a certain amount of white calcareous spots | 83.07 | 85.52 | 2.45 |
| Loess L11 | Brownish yellow sub-sandy soil, uniform, slightly dense, with occasional white calcareous spots | 85.52 | 86.26 | 0.74 |
| Paleosol S11 | Brownish yellow to reddish brown loam, uniform and dense, with a lot of white calcareous spots, and a small amount of calcium nodules in the upper part | 86.26 | 87.82 | 1.56 |
| Loess L12 | Brownish yellow sub-sandy soil, uniform, slightly dense, with a small amount of white calcareous spots | 87.82 | 88.54 | 0.72 |
| Paleosol S12 | Brownish yellow sandy loam, uniform, slightly dense, with a small amount of white calcareous spots | 88.54 | 89.30 | 0.76 |
| Loess L13 | Grayish yellow sub sandy soil, uniform, slightly dense, with a small amount of white calcareous spots, and a large amount of calcium nodules locally | 89.30 | 92.08 | 2.78 |
| Paleosol S13 | Red-brownish loam, uniform and dense, with a lot of white calcareous spots | 92.08 | 93.90 | 1.82 |
| Loess L14 | Brownish yellow to light yellow sub sandy soil, uniform and loose, with a small amount of white calcareous spots in the lower part | 93.90 | 95.05 | 1.15 |
| Paleosol S14 | Reddish brown loam, uniform and dense, with a large amount of white calcareous spots on the upper part and a small amount of white calcareous spots on the bottom | 95.05 | 96.43 | 1.38 |
| Loess L15 | Gray-yellowish to brown-yellowish silty fine sand, loose, containing some calcium nodules | 96.43 | 103.09 | 6.66 |
| Paleosol S15 | Brownish yellow to reddish brown loam, uniform and dense, containing a small amount of white calcareous spots, and calcium nodules are occasionally seen at the bottom | 103.09 | 104.78 | 1.69 |
| Loess L16 | Grayish yellow silty fine sand to sub sandy soil, locally containing calcium nodules | 104.78 | 107.48 | 2.70 |
| Paleosol S16 | Red-brownish loam, uniform and dense, with a few white calcareous spots | 107.48 | 108.74 | 1.26 |
| Loess L17 | Grayish yellow to brownish yellow sub sandy soil, containing calcium nodules locally, and a small amount of white calcium spots | 108.74 | 109.52 | 0.78 |
| Paleosol S17 | Red-brownish loam, uniform and dense, with a few white calcareous spots and calcium nodules | 109.52 | 110.50 | 0.98 |
| Loess L18 | Brownish yellow sub-sandy soil, containing a lot of white calcium spots and small calcium nodules | 110.50 | 111.24 | 0.74 |
| Paleosol S18 | Reddish brown loam, with occasional calcium nodules | 111.24 | 112.18 | 0.94 |
| Loess L19 | Brownish yellow sub-sandy soil, with a lot of white calcareous spots, and locally containing calcium nodules | 112.18 | 113.20 | 1.02 |
| Paleosol S19 | Reddish brown loam, uniform and dense, with a few white calcareous spots | 113.20 | 113.97 | 0.77 |
| Loess L20 | Grayish yellow silty fine sand, uniformly and loose, locally containing calcium nodules | 113.97 | 114.91 | 0.94 |
| Paleosol S20 | Brownish yellow to reddish brown loam, with a few white calcareous spots | 114.91 | 116.39 | 1.48 |
| Loess L21 | Brownish yellow sub-sandy soil, with a lot of white calcareous spots | 116.39 | 117.09 | 0.70 |
| Paleosol S21 | Red-brownish loam, with a few white calcareous spots | 117.09 | 117.93 | 0.84 |
| Loess L22 | Grayish yellow sub sandy soil, containing a small amount of white calcareous spots, and locally containing calcium nodules | 117.93 | 118.74 | 0.81 |
| Paleosol S22 | Reddish brown loam, with a few white calcareous spots | 118.74 | 119.86 | 1.12 |
| Loess L23 | Brownish yellow sub-sandy soil, with a small amount of white calcareous spots, and locally containing calcium nodules | 119.86 | 120.88 | 1.02 |
| Paleosol S23 | Reddish brown loam, uniform and dense, containing a lot of white calcareous spots, and locally containing calcium nodules | 120.88 | 121.86 | 0.98 |
| Loess L24 | Grayish yellow to brownish yellow sandy soil, uniform, slightly dense, containing calcium nodules | 121.86 | 128.35 | 6.49 |
| Paleosol S24 | Reddish brown loam, uniform and dense, with a few white calcareous spots | 128.35 | 129.29 | 0.94 |
| Loess L25 | Grayish yellow to brownish yellow sub sandy soil, with calcium nodules in local layers | 129.29 | 131.62 | 2.33 |
| Paleosol S25 | Brownish red loam, uniform and dense, containing a small amount of white calcareous spots, and locally containing calcium nodules | 131.62 | 133.44 | 1.82 |
| Loess L26 | Brownish yellow sub-sandy soil, with a small amount of white calcareous spots, and local layers containing calcium nodules | 133.44 | 135.76 | 2.32 |
| Paleosol S26 | Reddish brown loam, uniform and dense, with a few white calcareous spots and a few calcium nodules | 135.76 | 137.07 | 1.31 |
| Loess L27 | Grayish yellow sub sandy soil, with a lot of calcium nodules | 137.07 | 140.07 | 3.00 |
| Paleosol S27 | Reddish brown loam, uniform and dense, with a few white calcareous spots | 140.07 | 141.49 | 1.42 |
| Loess L28 | Grayish yellow to brownish yellow sub sandy soil, homogeneous, slightly dense, with a small amount of white calcareous spots, and local layer is calcium nodule layer | 141.49 | 144.25 | 2.76 |
| Paleosol S28 | Reddish brown loam, uniform and dense, containing a small amount of white calcareous spots and calcium nodule layer | 144.25 | 145.89 | 1.64 |
| Loess L29 | Gray-yellowish sandy loam with a calcium nodule layer | 145.89 | 148.42 | 2.53 |
| Paleosol S29 | Brownish red loam, uniform and dense, with white calcareous spots in the middle and lower part | 148.42 | 149.42 | 1.00 |
| Loess L30 | Grayish yellow sandy soil, with calcium nodule layer | 149.42 | 150.40 | 0.98 |
| Paleosol S30 | Reddish brown loam to clay, uniform and dense, with a few white calcareous spots and a few calcium nodules | 150.40 | 151.24 | 0.84 |
| Loess L31 | Brownish yellow sub-sandy soil, with a small amount of calcium nodules | 151.24 | 152.44 | 1.20 |
| Paleosol S31 | Brownish yellow to brownish red loam, uniform and dense, containing white calcareous spots and calcium nodules locally | 152.44 | 153.33 | 0.89 |
| Loess L32 | Gray-yellowish to brown yellowish sandy loam, interspersed with red-brownish loam, with locally observed calcium nodule layer | 153.33 | 163.29 | 9.96 |
| Paleosol S32 | Brown-reddish loam, uniform and dense, locally containing calcium nodules | 163.29 | 164.59 | 1.3 |
| Loess L33 | Gray-yellowish silty fine sand and sandy loam with abundant calcium nodules | 164.59 | 167.03 | 2.44 |
| Neogene N | Brown-reddish clay, homogeneous, relatively dense, with abundant calcium nodules | 167.03 | 200.42 | 33.39 |
2.2 Sampling and testing methods
2.2.1 Paleomagnetism
In this study, 139 (∼184 m) U-channel strip samples (specific U-shaped non-magnetic paleomagnetic sampling slot, 2 cm × 2 cm × 150 cm) were collected from the FX core, which were then tested at the State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences. The 2G-755R U-channel superconducting magnetometer system (2G Enterprise) was used to perform the alternating demagnetization test on the core samples collected from the FX core with a 2 cm resolution, measurement range of 2.0 × 10−12–2.0 × 10−4 Am2 and sensitivity of 2.0 × 10−12 Am2. The magnetic inclination and magnetic declination results were obtained after data processing (Figure 2).
Orthogonal vector projections of progressive demagnetization for selected samples of FX core.
2.2.2 Magnetic susceptibility measurements
The FX cores were sampled at 2 cm intervals to obtain 10,028 samples in total. About 10 g of air-dried samples were weighed and ground until the diameter size of the soil grains was <2 mm. The magnetic susceptibility of the samples was measured using a Bartington (UK) MS2 magnetic susceptibility system. The low frequency (0.47 Hz) (χ lf) and high frequency (4.7 Hz) (χ hf) magnetic susceptibilities of each sample were measured in the absence of interfering magnetic fields, and the average value was calculated from three consecutive measurements each.
2.2.3 Particle size analysis
Samples are the same as the ones used for the magnetic susceptibility. Specifically, 2–4 g of air-dried soil samples were weighed and placed in a beaker; subsequently, 10 mL H2O2 and 10 mL 10% HCl were added in the beaker. The mixture was heated on a heating plate to facilitate complete removal of organic matter and carbonates. Later, distilled water was added and the mixture was kept still for 24 h. The resultant supernatant was then extracted and mixed with 10 mL of a dispersing agent (0.05 mol/L sodium hexametaphosphate). After sonication, the particle size frequency distribution was measured using a Malvern Mastersizer 2000 laser particle size analyzer, with a measurement range of 0.02–2,000 µm. The particle size was analyzed to determine the percentage, median, and mean particle sizes of the various deposit components for different particle size levels.
2.2.4 LOI analysis
Here, 3,456 samples were selected at 4 cm intervals for LOI measurements. The crucibles and samples were dried at 105°C. Subsequently, the samples were ground, sieved, and burnt at 550 and 950°C for 2 h. Further, the LOI percentages related to the organic matter and carbonate contents were calculated, and the LOI–core depth variation curves were plotted.
3 Results
3.1 Paleomagnetism
Figure 2 shows the orthogonal vector projections of progressive demagnetization for selected samples of FX borehole. All samples were demagnetized in an alternating field of 0–80 mT, and the demagnetization data were fitted with the principal component analysis method to obtain the characteristic remanence direction of the samples (Figure 3) [24,25]. Data processing was done using the PaleoMag program [26].
Results of paleomagnetic polarity measurement of FX core.
The drill cores cannot be positioned relative to north; therefore, the magnetic declination of the characteristic remanence has no geological relevance. Thus, the magnetic polarity of the samples can only be determined using their magnetic inclination. According to the principles of paleomagnetism [27] and considering the errors arising from sample collection and testing, raw data were processed to remove single negative and positive polarity data from the continuous positive and negative polarity bands, respectively; additionally, data considered as transient polarity reversal events and containing less than three continuous polarity points were eliminated. Further, based on the actual depth of the samples, we plotted the magnetic inclination angle versus depth of the FX core. The magnetic polarity sequence of the FX core was established (Figure 3) by referring to the geomagnetic polarity time scale [28] and the comparison results of the magnetic susceptibility with the deep-sea oxygen isotope curve [29]. The paleomagnetic result is showing considerable differences to the global magnetic polarity time scale. The FX core magnetic polarity sequence recorded a total of 51 magnetic polarity bands, including 26 positive and 25 negative bands. The identifiable negative polarity drift events in the Brunhes positive polarity section included Mono Lake, Laschamp, Blake, Pringle Falls, Calabrian Ridge, and Stage 17, while other possible negative polarity drift events needed to be confirmed. The identifiable positive polarity drift events in the Matuyama negative polarity section were Jaramillo, Cobb Mt., Bjorn, Gardar, Gilsa, Olduvai, and Reunion, and other possible positive polarity drift events also needed to be confirmed. In addition, considering that no data of the core are screened, it was far-fetched that a few magnetic dip changes were defined as polarity events. In the lower part of FX core polarity records there were more polarity reversals than in the standard magnetostratigraphic time scale, which may be due to the abundance of calcium nodules and partial disturbance during accumulation in these layers.
According to the magnetic susceptibility of the loess–paleosols, the distribution positions of several important paleomagnetic boundaries in this core were identified to be virtually the same as those previously identified in other loess profiles. Specifically, the B/M boundary was in the middle and lower parts of L8, and top boundary of the Jaramillo event was in L10 and its bottom boundary was in L12. Moreover, the top and bottom boundaries of the Olduvai event were in L25 and lower part of S26, and the M/G boundary was in the middle and lower parts of L33.
Based on the abovementioned magnetic polarity comparisons, the magnetic chronological frame of the cores obtained from the FX core was established, and the variation of the sedimentation rate of the entire sequence was calculated (Figure 4). Based on the results, the sedimentation rate of the FX core could be divided as follows: The fastest sedimentation rate was observed in the cores in the first 84 m (average sedimentation rate = 0.0859 m/ka), followed by the relatively fast sedimentation rate in the cores in 84–131 m (average sedimentation rate = 0.0589 m/ka), while the sedimentation rate in the core after 131 m was relatively slow (average sedimentation rate = 0.0444 m/ka).
Magnetostratigraphic age model of FX core.
3.2 Magnetic susceptibility
Compared with that of the standard loess strata in Luochuan of central Loess Plateau and Lingtai of Gansu Province [8,10,30], the loess accumulation sequence in the Luochuan loess tableland area as shown by the magnetic susceptibility curves of the FX core exhibited unique characteristics (Figure 5). Moreover, compared with loess, the S1–S15 layers had precise and distinct curves with prominent peaks and layers, particularly the S1–S5 layers, with the highest value observed in the well-developed S5 paleosol, while the difference between the magnetic susceptibility values of paleosol and loess in the remaining layers decreased. The S1 paleosol had a large thickness, a brownish-red color, and a prominent prismatic structure, whereas the S2 layer included two paleosol layers and showed a magnetic susceptibility curve of two peaks. The S5 composite paleosol layer had three distinct overlapping paleosol layers. The top paleosol layer was the most developed, characterized by “three red stripes” and a prominent thickness of 4.94 m in the stratum. Comparatively, the paleosol layer inside Malan loess was weakly developed, with a relatively stable magnetic susceptibility. Some loess layers, such as L1, L9, L24, L27, and L32, consisted of weak paleosol layers. Among these, L24 and L32 layers were highly thick, while L5, L9, and L15 layers displayed sandy accumulation.
Comparison between FX core magnetic susceptibility and standard loess strata of Luochuan and Lingtai sections.
The test results indicated that the magnetic susceptibility ranged from 17 to 274 × 10−8 m3/kg, with the maximum value of low frequency observed in S5 (273.80 × 10−8 m3/kg), followed by S13 (248.75 × 10−8 m3/kg), S31 (234.90 × 10−8 m3/kg), S32 (217.05 × 10−8 m3/kg), and S23 (216.15 × 10−8 m3/kg), while the minimum value was observed in L15 (17.15 × 10−8 m3/kg). In the paleosol layers with strong loamification, the magnetic susceptibility values were relatively high and showed high and narrow peaks in the magnetic susceptibility curve. Moreover, the values indicated a warmer and more humid climate with a relatively short duration. In the paleosol layers with low loamification, the difference between the magnetic susceptibility values of loess and sand layers were not significant, but some variation in the magnetic susceptibility could still be observed. Further, the magnetic susceptibility of the loess layers was extremely low and exhibited a “wide and flat valley curve,” indicating a dry and cold climate. The variations in the magnetic susceptibility revealed multiple cycles of climate changes in the Quaternary period in the Luochuan loess tableland. Moreover, the magnetic susceptibility exhibited minor variations in the early Meso-Pleistocene, and the amplitude of the magnetic susceptibility became larger in the late Meso-Pleistocene and Neo-Pleistocene.
3.3 Particle size
The median particle size parameter was selected to reflect the overall coarseness of the particles of the entire sample. Additionally, 4 and 63 μm particle sizes were set as the criteria to classify clay, and silt and sand, respectively. The variation curve of the particle size of the loess–paleosol sequence in the FX core (Figure 6) indicated that the average silt content was the highest (∼80%) in the entire sample profile, followed by the clay (∼15%) and sand contents (∼5%). Thus, the overall particle size distribution was relatively concentrated and largely dominated by the silt fraction.
Grain size curves of different particle sizes in FX core.
The loess and paleosol layers displayed different particle size compositions, with the former containing relatively high silt and sand contents and low clay content, while the latter exhibiting an opposite trend. The median particle size of the paleosol layers was fine (7–28 μm), while the median particle size of the loess layers was comparatively coarser (12–45 μm). These variations in the particle size composition indicated that the loess layers were formed in a dry, cold, and relatively harsh climate, while the paleosol layers were found in a relatively warm and wet climate [31–33].
Vertically, the particle size content of the loess–paleosol layers varied markedly. The overall characteristics revealed that the upper parts of Malan and Lishi loess were comparatively coarse and the lower part of Lishi loess and Wucheng loess were finer, indicating an increasingly dry and cold climatic environment spanning 2.6 million years after the formation of the loess horizon in the Quaternary period.
3.4 LOI
LOI is the percentage loss of a mass of a sample under certain high temperatures in relation to the total mass [34]. Samples contain multiple components, and some components undergo physical and chemical processes, such as volatilization, combustion, and decomposition under certain temperature conditions. Therefore, LOI can indicate the content of certain components in the sample. Based on previous research [35–37], the high temperatures of LOI were set to 550°C for 2 h and 950°C for 2 h to reveal the organic matter and carbonate content of the deposits, respectively.
The LOI variation curve of the loess–paleosol sequence of the FX core (Figure 7) suggested that the organic matter content of the core profile varied between 1.79 and 5.21%, with the organic matter content of the paleosol being generally higher than that of the loess layer. Moreover, the carbonate content ranged from 0.60 to 17.67%, with the carbonate content in the loess layer being higher than that in the paleosol layer.
LOI curves in FX core.
4 Discussion
4.1 Paleoclimatic significance revealed by magnetic susceptibility
Several studies have concluded [38–42] that the magnetic susceptibility of loess indicated variations in the sedimentary environment, which has been extensively recognized and applied as a suitable proxy for the changes in the paleoclimatic environment. The variations in the magnetic susceptibility reflect the variations in the pedogenetic intensity, which effectively indicates the changes in the summer monsoon strength. Strong summer monsoons result in increased precipitation and paleosol development, thereby increasing the magnetic susceptibility. Conversely, weak summer monsoons result in reduced precipitation and loess accumulation, thus, reducing the magnetic susceptibility.
Furthermore, the magnetic susceptibility curve of the FX drill cores indicated rapid changes in the monsoon conditions in the plateau during glacial–interglacial transitions (Figure 5). High magnetic susceptibility occurred in S1, S2, S3, S4, S5, S6, S7, S8, S9, S12, and S13 and other paleosol developmental stages. Comparison of the standard loess time series in Luochuan revealed that during 129–71, 254–188, 334–279, 428–385, 576–471, 670–658, 748–706, 788–760, 883–853, 1,000–967, and 1,120–1,061 ka, the paleosols developed in Fu County, thus, indicating relatively warm and humid periods. The thick layers of loess/sand accumulations reflected cold and arid climates, occurring from ca 2.0 Ma, along with extreme climates during 71–12, 188–130, 380–334, 471–428, 658–576, 853–788, 1,273–1,265, and 1,727–1,640 ka. Moreover, sand accumulation occurred in L5, L9, and L15, indicating strong winds and a relatively dry and cold climate during the abovementioned periods.
4.2 Paleoclimatic significance based on particle size
Particle size is the most fundamental physical feature of sediments and is primarily controlled by the range of the source area, deposition dynamics, and weathering [6,43]. According to wind and sand dynamics, threshold wind velocity and particle size are directly related; the higher the wind velocity, coarser the material can be transported. Therefore, the median particle size can indicate the strength of wind dynamics. Previous studies on loess in Northern China have used the median particle size as a proxy indicator of winter wind [44–46], which transports coarser particles at high intensity and relatively finer particles at low intensity.
L1, L2, L3, L4, L5, L6, L9, L15, L29, and L33 display coarse particles and were speculated to be formed under a dry, cold, and harsh climate. Particularly, L5, L9, and L15 were prominent and indicated an excessively dry, cold, and harsh environment. Further, the particle size composition of the paleosol layers was relatively stable, suggesting that they were formed under relatively similar environmental conditions. The silt and sand contents in the paleosol layers decreased from the top to bottom layer, while the clay content increased. This indicated that the climate became drier, and the temperature gradually decreased during the formation of loess–paleosol sequences in this area since 2.6 Ma.
The fine particle fraction represents dust deposition owing to the weak monsoon circulation; additionally, its particle size indicates changes in the intensity of the near-surface airflow. The fine fraction potentially originates from the background dust controlled by the free high altitude westerly circulation, and the variations in the particle size characteristics indicate variations in the intensity of the high altitude airflow [47–49]. Overall, the core showed several strong fluctuations in the coarse fraction (>63 µm) content, which also indicated that the low altitude monsoon circulation strengthened during the glacial periods and weakened during the interglacial periods. Subsequently, the coarse fraction content decreased from the top to bottom layers, thus, demonstrating a clear strong near-surface circulation since 2.6 Ma.
4.3 Paleoclimatic significance based on LOI
Organic matter is produced under specific biological and climatic conditions, and its content and properties depend on the environmental conditions [50]. A warm and humid climate promotes the growth of natural vegetation and organic matter production and accumulation. Contrastingly, a dry and cold climate results in poor growth of natural vegetation and nutrient-deficient soil organic matter. The loess formation process can be viewed as the primary stage of the pedogenetic process to a certain extent. Therefore, organic matter production and accumulation in loess is likely to resemble these processes in soil. Hence, the high-temperature LOI of loess–paleosol at 550°C, to some extent, reflects the variation characteristics of the climate and environment during the loess accumulation period [51,52].
Carbonate, as a key climate proxy, is extensively used to understand the climatic conditions during loess formation [53]. Our study indicated that during the formation of paleosols, the moisture conditions were adequate under warm and humid climatic conditions, and dust materials were subject to strong weathering and leaching effects. Moreover, CaCO3 underwent a strong leaching process, which caused the leaching of CaCO3 from the paleosol and subsequent deposition in the underlying loess layer, consequently forming calcium nodules. In the loess layers, carbonates were preserved well and their contents were relatively high, owing to the dry and cold climate during their formation periods; additionally, the weathering effect was weak. Therefore, high carbonate content indicated a cold and dry climate, while lower carbonate content reflected a more humid climate [54,55].
The peak values shown in the W550 curve of Figure 6 revealed that the paleosols were subjected to strong biological and chemical weathering during the formation period, thus, indicating a hot and humid climate. Conversely, the low values reflected a dry and cold climate during the loess accumulation period. The low values displayed by the W950 curve indicated highly reduced carbonate content due to strong leaching under a relatively humid climate. The peak values at the other end of the curve indicated dryer and colder climate with relatively weaker weathering, during which carbonates were appropriately preserved (or precipitated below paleosols), thereby resulting in higher carbonate content. This alternation of high and low values reflects the basic cycles of paleoclimate changes of the Loess Plateau.
To sum up, through the research on the lithostratigraphy, chronology, magnetic susceptibility, particle composition, and high-temperature LOI of the loess–paleosol series in FX core, the change curves of climate substitute indicators such as magnetic susceptibility, grain size, and LOI on the profile have been established. The analysis results show that these indirect climate indicators can reflect the paleoclimate changes in the loess accumulation period with different meanings, and almost all of these indicator curves show a periodic change rule, revealing that a warmer climate during the paleosol formation period, experienced strong biochemical weathering, while a relatively dry and cold climate during the loess accumulation period.
5 Conclusions
In this study, paleomagnetic chronological analysis, along with particle size and magnetic susceptibility tests were performed to reveal the loess stratigraphy characteristics in Luochuan loess tableland in Fu County, Shaanxi Province, China. The main conclusions of the study are as follows:
The FX core reveals a Quaternary loess archive of 167.03 m in the area, which included 33 developed layers of loess–paleosol sequences, including 37 loess–paleosol assemblages, among which S2 and S9 each included two superimposed soil layers, while S5 included three composite soil layers.
The loess–paleosol assemblages represented 37 warm and cold climate cycles since 2.6 Ma. Specifically, 10 paleoclimate cycles (S23–L33) occurred for 1 million years from 2.6 to 1.6 Ma, 15 climate cycles (S9–L23) occurred for 700,000 years from 1.6 to 0.8 Ma, and 12 loess–paleosol cycles (L9–S0) from 0.8 Ma to the present.
Furthermore, the amplitude of the magnetic susceptibility of loess–paleosols revealed that among the 37 large warm periods from 2.6 Ma to the present, the early (ca 2.6–1.15 Ma BP) and late (ca 0.73 Ma BP to the present) warm periods were warmer and more humid than the middle warm period (ca 1.15–0.73 Ma BP). Moreover, among the 37 cold periods, the climate of the middle (ca 1.15–0.73 Ma BP) cold period was warmer and more humid than that of the early (∼2.6–1.15 Ma BP) and late (∼0.73 Ma BP to the present) cold periods. These trends suggested that the climate was relatively drier in the 37 cold periods than in the warm periods.
Further analysis of climate proxies, such as magnetic susceptibility, particle size, and LOI revealed that the alternating occurrence of multiple loess and paleosol layers in the FX core and the corresponding multiple transitions between dry–cold and warm–wet climates represented the major characteristics of geological events in the Luochuan loess tableland of the Loess Plateau. Specifically, the paleosol layers indicated relatively warm and wet climatic conditions, while the sandy loess layers with coarse particle deposition reflected cold and dry climatic events.
Acknowledgements
We would like to thank Editage (www.editage.cn accessed on 2 October 2022) for English language editing.
-
Funding information: This study was financially supported by the Basic Scientific Research Project (Grant No. SK201403) and China Geological Survey Project (Grant No. DD20190433, 1212011120047).
-
Conflict of interest: The authors declare no conflict of interest.
-
Data availability statement: The data presented in this study are available on request from the corresponding author.
References
[1] Guo ZT, Ruddiman WF, Hao QZ, Wu HB, Qiao YS, Zhu RX, et al. Onset of Asian desertification by 22 Myr ago inferred from loess deposits in China. Nature. 2002;416:159–63.10.1038/416159aSearch in Google Scholar PubMed
[2] Zhai XM, Lu HY, Li LP, Tang GA. Landscape evolution and erosion of Luochuan tableland (central Chinese Loess Plateau) at different time scales. Quat Sci. 2012;32(5):839–48.Search in Google Scholar
[3] An ZS, Kukla GJ, Porter SC, Xiao JL. Magnetic susceptibility evidence of monsoon variation on the Loess Plateau of central China during the last 130,000 years. Quat Res. 1991;36(1):29–36.10.1016/0033-5894(91)90015-WSearch in Google Scholar
[4] An ZS, Colman SM, Zhou WJ, Li XQ, Brown ET, Jull AJ, et al. Interplay between the Westerlies and Asian monsoon recorded in Lake Qinghai sediments since 32 ka. Sci Rep. 2012;2:619.10.1038/srep00619Search in Google Scholar PubMed PubMed Central
[5] Liu TS. Material composition and structure of loess. Beijing, China: Science Press; 1966.Search in Google Scholar
[6] Liu TS. Loess and environment. Beijing, China: Science Press; 1985.Search in Google Scholar
[7] Heller F, Liu TS. Magnetostratigraphical dating of loess deposits in China. Nature. 1982;300:431–3.10.1038/300431a0Search in Google Scholar
[8] An ZS, Kukla G, Liu TS. Loess stratigraphy in Luochuan of China. Quat Sci. 1989;9(2):155–68.Search in Google Scholar
[9] Sun JM, Liu TS. Pedostratigraphic subdivision of the loess–paleosol sequences at Luochuan and a new interpretation on the paloenvironmental significance of L9 and L15. Quat Sci. 2002;22(5):406–12.Search in Google Scholar
[10] Lu HY, Yang WF, Liu XD, An ZS. Orbital tuned time scale for loess deposit in Luochuan of central Chinese Loess Plateau. Acta Geophys Sin. 1998;41(6):804–10.Search in Google Scholar
[11] Liu WM, Zhang LY, Sun JM. High resolution magnetostratigraphy of the Luochuan loess–paleosol sequence in the central Chinese Loess Plateau. Chin J Geophys. 2010;53(4):888–94.Search in Google Scholar
[12] Wu ZR, Yuan BY, Gao FQ. Geological environment of loess at Luochuan Yuan, northern Shaanxi Province, China. Quat Sci. 1985;6(1):137–48.Search in Google Scholar
[13] Han JM, Jiang WY, Wu NQ, Guo ZT. Carbon and oxygen isotope compositions of carbonate concretions in loess part 1: oxygen isotope and paleotemperature. Quat Sci. 1995;15(2):130–8.Search in Google Scholar
[14] Han JM, Jiang WY, Lv HY, Wu NQ, Guo ZT. Carbon and oxygen isotope compositions of carbonate concretions in loess part 2: carbon isotope and paleo-aridity. Quat Sci. 1995;15(4):367–77.Search in Google Scholar
[15] Wen QZ, Diao GY, Jia RF, Zhou HY. Geochemical records of paleoclimate change in loess sections. Quat Sci. 1995;15(3):223–31.Search in Google Scholar
[16] Chen JA, Wan GJ, Zhang DD, Zhang F, Huang RG. Environmental records of lacustrine sediments in different time scales: sediment grain size as an example. Sci China (Ser D). 2004;47(10):954–60.10.1360/03yd0160Search in Google Scholar
[17] Chen J, Wang YJ, Ji JF, Chen Y, Lu HY. Rb/Sr variations and its climatic stratigraphical significance of a loess–paleosol profile from Luochuan, Shaanxi Province. Quat Sci. 1999;19(4):350–6.Search in Google Scholar
[18] Chen Y, Chen J, Ji JF, Liu LW, Lu HY. Whiteness intensity in Luochuan loess sequence (Shaanxi Province) and paleoclimate implications. Geol Rev. 2002;48(1):38–43.Search in Google Scholar
[19] Ding ZL, Sun JM, Yang SL, Liu TS. Preliminary magnetostratigraphy of a thick eolian red clay‐loess sequence at lingtai, the Chinese Loess Plateau. Geophys Res Lett. 1998;25(8):1225–8.10.1029/98GL00836Search in Google Scholar
[20] Lv HY, Wang YJ. A study on phytoliths in loess profile and paleoenvironmental evolution at Heimugou in Luochuan, Shaanxi Province since late Pleistocene. Quat Sci. 1991;11(1):72–84.Search in Google Scholar
[21] Chen XY, Wu NQ. Relatively warm-humid climate recorded by mollusk species in the Chinese Loess Plateau during MIS 3 and its possible forcing mechanism. Quat Sci. 2008;28(1):154–61.Search in Google Scholar
[22] Yang MS, Zhang HC, Lei GL, Zhang WX, Fan HF, Chang FQ, et al. Biomarkers in weakly developed paleosol (L1SS1) in the Luochuan loess section and reconstructed paleovegetation-environment during the interstade of the last glaciation. Quat Sci. 2006;26(6):976–84.Search in Google Scholar
[23] Xue GY, Zhou SQ, Chen Y, Li MJ, Chen HM, Quan LH. Significant spectrum evolution of the east Asian summer monsoon on the Loess Plateau during the last 2.6 Ma B.P. Clim Environ Res. 2007;12(1):63–73.Search in Google Scholar
[24] Zijderveld JDA. A. C. demagnetization of rocks: analysis of results. Dev Solid Earth Geophys. 2013;3:254–86.10.1016/B978-1-4832-2894-5.50049-5Search in Google Scholar
[25] Kirschvink JL. The least-squares line and plane and the analysis of palaeomagnetic data. Geophys J Int. 1980;62(3):699–718.10.1111/j.1365-246X.1980.tb02601.xSearch in Google Scholar
[26] Jones CH. User-driven integrated software lives: “PaleoMag” paleomagnetics analysis on the Macintosh. Comput Geosci. 2002;28(10):1145–51.10.1016/S0098-3004(02)00032-8Search in Google Scholar
[27] Qiang XK, An ZS, Song YG, Chang H, Sun YB, Liu WG, et al. New eolian red clay sequence on the western Chinese Loess Plateau linked to onset of Asian desertification about 25 Ma ago. Sci China Earth Sci. 2010;54:136–44.10.1007/s11430-010-4126-5Search in Google Scholar
[28] Gradstein F, Ogg J, Smith A. A geologic time scale. United Kingdom: Cambridge University Press; 2004.Search in Google Scholar
[29] Shackleton NJ, Berger A, Peltier WR. An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Trans R Soc Edinburgh Earth Sci. 1990;81:251–61.10.1017/S0263593300020782Search in Google Scholar
[30] Sun YB, Clemens SC, An ZS, Yu ZW. Astronomical timescale and palaeoclimatic implication of stacked 3.6-Myr monsoon records from the Chinese Loess Plateau. Quat Sci Rev. 2006;25:33–48.10.1016/j.quascirev.2005.07.005Search in Google Scholar
[31] Lu HY, An ZS. Paleoclimatic significance of grain size composition of loess in the loess plateau. Sci China (Ser D) . 1998;28(3):278–83.Search in Google Scholar
[32] Xu CQ, Jia YL, Liu Q, Ling CH, Wang PL. Grain-size characteristics of Quaternary sediments and its implication in the Poyang Lake region, northern Jiangxi. J Arid Land Resour Environ. 2016;30(9):104–8.Search in Google Scholar
[33] Wang HY, Pang JL, Huang CC, Zhou YL, Gao PK, Wang LB. Grain-size composition of Chafangcun loess–paleosol sequence in Shangdan Basin, Shaanxi and their significance. Acta Sci Nat Univ Sunyatseni. 2017;56(2):131–9.Search in Google Scholar
[34] Dean WE. Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition; comparison with other methods. J Sediment Petrol. 1974;44:242–8.10.1306/74D729D2-2B21-11D7-8648000102C1865DSearch in Google Scholar
[35] Hu CL, Ma YZ, Guo C, Li DD, Pei QM. Optimization of the experiment conditions for estimating organic matter content with loss-on-ignition method. Earth Environ. 2016;44(1):110–8.Search in Google Scholar
[36] Chen S, Chen Y, Zhao JQ, Zhang CL. Study on the relationship between soil loss on ignition and organic matter content. Sichuan Environ. 2021;40(5):17–21.Search in Google Scholar
[37] Zhang WH, Mu GJ. Precision control on measuring organic and carbonate content with loss on ignition method. Arid Land Geogr. 2007;30(3):455–9.Search in Google Scholar
[38] Balsam W, Ellwood B, Ji JF. Direct correlation of the marine oxygen isotope record with the Chinese Loess Plateau iron oxide and magnetic susceptibility records. Palaeogeogr Palaeoclimatol Palaeoecol. 2005;221(1–2):141–52.10.1016/j.palaeo.2005.02.009Search in Google Scholar
[39] Heller F, Liu TS. Magnetism of Chinese loess deposits. Geophys J Int. 1984;77(1):125–41.10.1111/j.1365-246X.1984.tb01928.xSearch in Google Scholar
[40] Maher BA. Palaeoclimatic records of the loess/palaeosol sequences of the Chinese Loess Plateau. Quat Sci Rev. 2016;154:23–84.10.1016/j.quascirev.2016.08.004Search in Google Scholar
[41] Peng WB, Nie JS, Wang Z, Qiang XK, Garzanti E, Pfaff K, et al. A major change in precipitation gradient on the Chinese Loess Plateau at the Pliocene—Quaternary boundary. J Asian Earth Sci. 2018;155(30):134–8.10.1016/j.jseaes.2017.10.031Search in Google Scholar
[42] Thomas EK, Clemens SC, Sun YB, Prell WL, Huang YS, Gao L, et al. Heterodynes dominate precipitation isotopes in the East Asian monsoon region, reflecting interaction of multiple climate factors. Earth Planet Sci Lett. 2016;455:196–206.10.1016/j.epsl.2016.09.044Search in Google Scholar
[43] Kong FB, Xu SJ, Han M, Chen HT, Miao XD, Kong XL, et al. Application of grain size endmember analysis in the study of dust accumulation processes: a case study of loess in Shandong Province, East China. Sediment Geol. 2021;416:105868.10.1016/j.sedgeo.2021.105868Search in Google Scholar
[44] Zhang XY. Source distributions, emission, transport, deposition of Asian dust and loess accumulation. Quat Sci. 2001;21(1):29–40.Search in Google Scholar
[45] Kang SG, Du JH, Wang N, Dong JB, Wang D, Wang XL, et al. Early Holocene weakening and mid- to late Holocene strengthening of the East Asian winter monsoon. Geology. 2020;48(11):1043–7.10.1130/G47621.1Search in Google Scholar
[46] Ueno Y, Hyodo M, Yang TS, Katoh S. Intensified East Asian winter monsoon during the last geomagnetic reversal transition. Sci Rep. 2019;9:9389.10.1038/s41598-019-45466-8Search in Google Scholar PubMed PubMed Central
[47] Wang ZD, Huang CC, Zhou YL, Pang JL, Zha XC. Characteristics of Holocene loess–palaeosol particle size composition and paleoclimatic significance in east Guanzhong, Shaanxi Province. Adv Earth Sci. 2018;33(3):293–304.Search in Google Scholar
[48] Wang LD, Yang TB, Liang Y, He Y, Chen J. Grain size characteristics in the loess–paleosol at Huining section and its signification to paleoclimate during Holocene. Arid Zone Res. 2016;33(6):1150–6.Search in Google Scholar
[49] Lu HY, Wang XY, Sun XF, Wang XY, Yi SW, Zhou YL, et al. Loess stratigraphy and palaeoclimate changes during Quaternary in northeastern Tibetan Plateau revealed by loess core. Quat Sci. 2007;27(2):230–241.Search in Google Scholar
[50] Liu ZT, Yu JQ, Zhang BH, Cai W, Zhang LS. Application of loss on ignition to the study of lake sediments and environmental changes. J Salt Lake Res. 2006;14(2):67–72.Search in Google Scholar
[51] Wu JJ, Chen SY, Yao M, Yao X, Yu SY, Chen YY. Loss-on-ignition record and historical environmental changes due to floodplain aggradation in lower Yellow River. J Anhui Norm Univ (Nat Sci). 2014;37(3):270–3.Search in Google Scholar
[52] Peng XZ, Jia RF. Distribution of organic matter in Xifeng and Duanjiapo loess sections and the paleo-environmental significance. Sci Geogr Sin. 2001;21(1):36–40.10.1007/BF02893780Search in Google Scholar
[53] Sui YZ. Study on environmental signification of main indexes of loess. J Desert Res. 2006;26(1):14–9.Search in Google Scholar
[54] Chen D, Xiao QL, Zhang YZ, Wang NL, Wang HY, Jia YN, et al. Weathering and pedogenesis of the Holocene aeolian loess–paleosol section and paleoclimate evolution in the Maqu reach in the Yellow River source area. Geogr Res. 2022;41(8):2277–94.Search in Google Scholar
[55] Shi JS, Shi YC, Ye H, Sun YM. Indication significance of “high-temperature loss on ignition” in loess accumulation sequence to palaeoclimate evolution. Geogr Territ Res. 2002;18(4):104–6.Search in Google Scholar
© 2023 the author(s), published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.
