Pedogeomorphology, geochronology and paleobotany of soil profiles obtained in the Kyichu River catchment, sourthern Tibet

Results of pedogeomorphological, geochronological and paleobotanical investigations are presented covering the last ca. 4,000 years. The study sites are located in the heavily degraded Kyichu River catchment around Lhasa at 3,600-4,600 m a.s.l. Repeatedly, colluvial sediments have been recorded overlying paleosols. These deposits can be divided into i) coarse-grained sediments with a high proportion of stones and boulders originating from alluvial fans and debris flows, ii) matrix supported sediments with some stones and boulders originating from mudflows or combined colluvial processes such as hillwash plus rock fall, and iii) fine-grained sediments originating from hill wash. The IRSL multi-level dating of profile QUG 1 points to a short-time colluvial sedimentation between 1.0 ± 0.1 and 0.8 ± 0.1 ka. In contrast, dated paleosols of profile GAR 1 (7,908 ± 99 and 3,668 ± 57 BP) encompass a first colluvial episode. Here, the upper colluvial sedimentation took place during several periods between 2.6 ± 0.3 and 0.4 ± 0.1 ka. For the first time in Tibet, a systematic extraction, determination and dating of charcoals from buried paleosols was conducted. The charcoals confirm the Late Holocene presence of juniper forests or woodlands in a now treeless, barren environment. A pollen diagram from Lhasa shows a distinct decline of pollen of the Jumperus-type around 4,140 ± 50 BP, which is interpreted as indicating a clearing of forests on the adjacent slopes. It is assumed that the environmental changes from forests to desertic rangelands since ca. 4,000 BP have been at least reinforced by humans.

Geochemistry of ODP Site 202-1233 and river sediments

Bulk sediment chemistry from three Chilean continental margin Ocean Drilling Program sites constrains regional continental erosion over the past 30,000 years. Sediments from thirteen rivers that drain the (mostly igneous) Andes and the (mostly metamorphic) Coast Range, along with existing rock chemistry datasets, define terrestrial provenance for the continental margin sediments. Andean river sediments have high Mg/Al relative to Coast-Range river sediments. Near 36°S, marine sediments have high-Mg/Al (i.e. more Andean) sources during the last glacial period, and lower-Mg/Al (less Andean) sources during the Holocene. Near 41°S a Ti-rich source, likely from coast-range igneous intrusions, is prevalent during Holocene time, whereas high-Mg/Al Andean sources are more prevalent during the last glacial period. We infer that there is a dominant ice-sheet control of sediment sources. At 36°S, Andean-sourced sediment decreased as Andean mountain glaciers retreated after ~17.6 ka, coincident with local oceanic warming and southward retreat of the Patagonian Forest and, by inference, westerly winds. At 41°S Andean sediment dominance peaks and then rapidly declines at ~19 ka, coincident with local oceanic warming and the earliest deglacial sea-level rise. We hypothesize that this decreased flux of Andean material in the south is related to rapid retreat of the marine-based portion of the Patagonian Ice Sheet in response to global sea-level rise, as the resulting flooding of the southern portion of the Central Valley created a sink for Andean sediments in this region. Reversal of the decreasing deglacial Mg/Al trend at 41°S from 14.5 to 13.0 ka is consistent with a brief re-advance of the Patagonian ice sheet coincident with the Antarctic Cold Reversal.

Pollen analysis of sediment cores from the mouth of the river Congo to Walvis Bay and Lüderitz

The distribution of pollen in marine sediments is used to record vegetation changes over the past 30,000 years on the adjacent continent. A transect of marine pollen sequences from the mouth of the river Congo (~5°S) to Walvis Bay and Lüderitz (~25°S) shows vegetation changes in Congo, Angola and Namibia from the last glacial period into the Holocene. The comparison of pollen records from different latitudes provides information about the latitudinal shift of open forest and savannahs (Poaceae pollen), the extension of lowland forest (rain forest pollen) and Afromontane forest (Podocarpus pollen), and the position of the desert fringe (pollen of Caryophyllaceae, Chenopodiaceae and Amaranthaceae). High Cyperaceae pollen percentages in sediments from the last glacial period off the mouth of the river Congo suggest the presence of open swamps rather than savannah vegetation in the Congo Basin. Pollen from Restionaceae in combination with Stoebe-type pollen (probably from Elytropappus) indicates a possible northwards extension of winter rain vegetation during the last glacial period. The record of Rhizophora (mangrove) pollen is linked to erosion of the continental shelf and sea-level rise. Pollen influx is highest off river mouths (10-2000 grains year**-1 cm**-2), close to the coast (300-6000 grains year**-1 cm**-2), but is an order of magnitude lower at sites situated far from the continent (<10 grains year**-1 cm**-2).

(Figure 2) Mean discharge per day (total and baseflow part) of the San Francisco River and daily precipitation at station ECPL (Planta), in 2007

We investigated controls on the water chemistry of a South Ecuadorian cloud forest catchment which is partly pristine, and partly converted to extensive pasture. From April 2007 to May 2008 water samples were taken weekly to biweekly at nine different subcatchments, and were screened for differences in electric conductivity, pH, anion, as well as element composition. A principal component analysis was conducted to reduce dimensionality of the data set and define major factors explaining variation in the data. Three main factors were isolated by a subset of 10 elements (Ca2+, Ce, Gd, K+, Mg2+, Na+, Nd, Rb, Sr, Y), explaining around 90% of the data variation. Land-use was the major factor controlling and changing water chemistry of the subcatchments. A second factor was associated with the concentration of rare earth elements in water, presumably highlighting other anthropogenic influences such as gravel excavation or road construction. Around 12% of the variation was explained by the third component, which was defined by the occurrence of Rb and K and represents the influence of vegetation dynamics on element accumulation and wash-out. Comparison of base- and fast flow concentrations led to the assumption that a significant portion of soil water from around 30 cm depth contributes to storm flow, as revealed by increased rare earth element concentrations in fast flow samples. Our findings demonstrate the utility of multi-tracer principal component analysis to study tropical headwater streams, and emphasize the need for effective land management in cloud forest catchments.

Discharge measurements and storm event chemistry at the San Francisco River station (Planta) in 2007-2008

To improve our knowledge of the influence of land-use on solute behaviour and export rates in neotropical montane catchments we investigated total organic carbon (TOC), Ca, Mg, Na, K, NO3 and SO4 concentrations during April 2007-May 2008 at different flow conditions and over time in six forested and pasture-dominated headwaters (0.7-76 km2) in Ecuador. NO3 and SO4 concentrations decreased during the study period, with a continual decrease in NO3 and an abrupt decrease in February 2008 for SO4. We attribute this to changing weather regimes connected to a weakening La Niña event. Stream Na concentration decreased in all catchments, and Mg and Ca concentration decreased in all but the forested catchments during storm flow. Under all land-uses TOC increased at high flows. The differences in solute behaviour during storm flow might be attributed to largely shallow subsurface and surface flow paths in pasture streams on the one hand, and a predominant origin of storm flow from the organic layer in the forested streams on the other hand. Nutrient export rates in the forested streams were comparable to the values found in literature for tropical streams. They amounted to 6-8 kg/ha/y for Ca, 7-8 kg/ha/y for K, 4-5 kg/ha/y for Mg, 11-14 kg/ha/y for Na, 19-22 kg/ha/y for NO3 (i.e. 4.3-5.0 kg/ha/y NO3-N) and 17 kg/ha/y for SO4. Our data contradict the assumption that nutrient export increases with the loss of forest cover. For NO3 we observed a positive correlation of export value and percentage forest cover.

Terrigenous biomarker proxies analysed in the Amazon River system

Paleoenvironmental studies based on terrigenous biomarker proxies from sediment cores collected close to the mouth of large river systems rely on a proper understanding of the processes controlling origin, transport and deposition of biomarkers. Here, we contribute to the understanding of these processes by analyzing long-chain n-alkanes from the Amazon River system. We use the dD composition of long-chain n-alkanes from river bed sediments from the Amazon River and its major tributaries, as well as marine core-top samples collected off northeastern South America as tracers for different source areas. The d13C composition of the same compounds is used to differentiate between long-chain n-alkanes from modern forest vegetation and petrogenic organic matter. Our d13C results show depleted d13C values (-33 to -36 per mil) in most samples, indicating a modern forest source for most of the samples. Enriched values (-31 to -33 per mil) are only found in a few samples poor in organic carbon indicating minor contributions from a fossil petrogenic source. Long-chain n-alkane dD analyses show more depleted values for the western tributaries, the Madeira and Solimões Rivers (-152 to -168 per mil), while n-alkanes from the lowland tributaries, the Negro, Xingu and Tocantins Rivers (-142 to -154 per mil), yield more enriched values. The n-alkane dD values thus reflect the mean annual isotopic composition of precipitation, which is most deuterium-depleted in the western Amazon Basin and more enriched in the eastern sector of the basin. Samples from the Amazon estuary show a mixed long-chain n-alkane dD signal from both eastern lowland and western tributaries. Marine core-top samples underlying the Amazon freshwater plume yield dD values similar to those from the Amazon estuary, while core-top samples from outside the plume showed more enriched values. Although the variability in the river bed data precludes quantitative assessment of relative contributions, our results indicate that long-chain n-alkanes from the Amazon estuary and plume represent an integrated signal of different regions of the onshore basin. Our results also imply that n-alkanes are not extensively remineralized during transport and that the signal at the Amazon estuary and plume includes refractory compounds derived from the western sector of the Basin. These findings will aid in the interpretation of plant wax-based records of marine sediment cores collected from the adjacent ocean.

(Tables 1-3) Water chemistry of cloud forest streams at baseflow conditions, Rio San Francisco, Ecuador

We investigated controls on the water chemistry of a South Ecuadorian cloud forest catchment which is partly pristine, and partly converted to extensive pasture. From April 2007 to May 2008 water samples were taken weekly to biweekly at nine different subcatchments, and were screened for differences in electric conductivity, pH, anion, as well as element composition. A principal component analysis was conducted to reduce dimensionality of the data set and define major factors explaining variation in the data. Three main factors were isolated by a subset of 10 elements (Ca2+, Ce, Gd, K+, Mg2+, Na+, Nd, Rb, Sr, Y), explaining around 90% of the data variation. Land-use was the major factor controlling and changing water chemistry of the subcatchments. A second factor was associated with the concentration of rare earth elements in water, presumably highlighting other anthropogenic influences such as gravel excavation or road construction. Around 12% of the variation was explained by the third component, which was defined by the occurrence of Rb and K and represents the influence of vegetation dynamics on element accumulation and wash-out. Comparison of base- and fast flow concentrations led to the assumption that a significant portion of soil water from around 30 cm depth contributes to storm flow, as revealed by increased rare earth element concentrations in fast flow samples. Our findings demonstrate the utility of multi-tracer principal component analysis to study tropical headwater streams, and emphasize the need for effective land management in cloud forest catchments.