An ecohydrological approach to predicting regional woody species distribution patterns in dryland ecosystems

This paper presents a quantitative ecohydrological framework for predicting regional distribution patterns of woody species in dryland ecosystems. The framework is based on an existing stochastic model for the daily mass balance of water that represents the interactions between soils, climate, and vegetation. Individual species selection is based on an optimality trade-off hypothesis, which states that dryland vegetation patterns are constrained by maximization of water use and simultaneous minimization of water stress. The relative importance of water use and stress avoidance to the overall fitness of three Acacia species is determined from the heterogeneous basin, the Upper Ewaso Ng’iro river basin, of the central Kenya highlands. The model results indicate that overall fitness is more strongly influenced by water use than stress avoidance but that consideration of both stress avoidance and water use is critical to predicting basin-scale patterns of species distribution. We identify a linear trend in the frequency and intensity of storms with the same annual total using a basin-wide gauge precipitation dataset. After calibration, we apply the basin average linear trends in time for average rain per storm and storm arrival rates. The model results indicate the upslope migration of two species, Acacia tortilis and Acacia xanthophloea to areas with higher total rainfall. Lastly, we explore the modeled changes of species cover in the basin influenced by changes in rainfall total holding growing season rainfall variability constant and changes in growing season rainfall variability holding total rainfall constant. We find that changes in dryland species distribution patterns and relative abundance may be as sensitive to growing season rainfall variability as they are to changes in total rainfall amounts.     

Development of a modelling methodology for simulation of long-term morphological evolution of the southern Baltic coast

The Darss–Zingst peninsula at the southern Baltic Sea is a typical wave-dominated barrier island system which includes an outer barrier island and an inner lagoon. The formation of the Darss–Zingst peninsula dates back to the Littorina Transgression onset about 8,000 cal BP. It originated from several discrete islands, has been reshaped by littoral currents, wind-induced waves during the last 8,000 years and evolved into a complex barrier island system as today; thus, it may serve as an example to study the coastal evolution under long-term climate change. A methodology for developing a long-term (decadal-to-centennial) process-based morphodynamic model for the southern Baltic coastal environment is presented here. The methodology consists of two main components: (1) a preliminary analysis of the key processes driving the morphological evolution of the study area based on statistical analysis of meteorological data and sensitivity studies; (2) a multi-scale high-resolution process-based model. The process-based model is structured into eight main modules. The two-dimensional vertically integrated circulation module, the wave module, the bottom boundary layer module, the sediment transport module, the cliff erosion module and the nearshore storm module are real-time calculation modules which aim at solving the short-term processes. A bathymetry update module and a long-term control function set, in which the ‘reduction’ concepts and technique for morphological update acceleration are implemented, are integrated to up-scale the effects of short-term processes to a decadal-to-centennial scale. A series of multi-scale modelling strategies are implemented in the application of the model to the research area. Successful hindcast of the coastline change of the Darss–Zingst peninsula for the last 300 years validates the modelling methodology. Model results indicate that the coastline change of the Darss–Zingst peninsula is dominated by mechanisms acting on different time scales. The coastlines of Darss and the island of Hiddensee are mainly reshaped by long-term effects of waves and longshore currents, while the coastline change of the Zingst peninsula is due to a combination of long-term effects of waves and short-term effects caused by wind storms.     

U–Pb dating of calcite–aragonite layers in speleothems from hominin sites in South Africa by MC-ICP-MS

Speleothems are found in association with hominin fossil-bearing cave deposits in South Africa and can be used to provide valuable chronological constraints. Such material is generally too old for U–Th dating and, although U–Pb geochronology presents a suitable alternative, bulk U concentrations are typically too low (<0.05 μg/g) to provide useful ages. For this reason, we used a simple non-invasive beta-scanner imaging screening step to identify U-rich (≥1 μg/g) domains that could be analyzed with MC-ICP-MS techniques to provide U–Pb ages. We demonstrate the technique using samples from Sterkfontein cave that exhibit infrequent <1 cm-thick layers with U concentrations ≥1 μg/g. Relict aragonite needles are found exclusively in these U-rich layers. We analyzed material from the same flowstone suite as Walker et al. (2006) and obtained a U–Pb age of ～2.3 Ma that agrees well with their estimate of 2.24 ± 0.09 Ma. We also obtained similar U–Pb (0.164 ± 0.026 to 0.200 ± 0.052 Ma) and U–Th (0.148 ± 0.003 Ma) ages for another sample exhibiting U-rich layers. We recognize that open-system behaviour during the partial transformation of aragonite to calcite is a potential problem and argue, on the basis of geochemistry and age consistencies, that recrystallization took place rapidly after speleothem formation and did not significantly affect the U–Pb ages.     

Silicate‐oxide mineral chemistry of mafic–ultramafic rocks as an indicator of the roots of an island arc: The Chilas Complex,Kohistan (Pakistan)

The Chilas Complex is a major lower crustal component of the Cretaceous Kohistan island arc and one of the largest exposed slices of arc magma chamber in the world. Covering more than 8000 km², it reaches a current tectonic width of around 40 km. It was emplaced at 85 Ma during rifting of the arc soon after the collision of the arc with the Karakoram plate. Over 85% of the Complex comprises homogeneous, olivine‐free gabbronorite and subordinate orthopyroxene–quartz diorite association (MGNA), which contains bodies of up to 30 km² of ultramafic–mafic–anorthositic association (UMAA) rocks. Primary cumulate textures, igneous layering, and sedimentary structures are well preserved in layered parts of the UMAA in spite of pervasive granulite facies metamorphism. Mineral analyses show that the UMAA is characterized by more magnesian and more aluminous pyroxene and more calcic plagioclase than those in the MGNA. High modal abundances of orthopyroxene, magnetite and ilmenite (in MGNA), general Mg–Fe–Al spatial variations, and an MFA plot of whole‐rock analyses suggest a calc‐alkaline origin for the Complex. Projection of the pyroxene compositions on the Wo–En–Fs face is akin to those of pyroxenes from island arcs gabbros. The presence of highly calcic plagioclase and hornblende in UMAA is indicative of hydrous parental arc magma. The complex may be a product of two‐stage partial melting of a rising mantle diaper. The MGNA rocks represent the earlier phase melting, whereas the UMAA magma resulted from the melting of the same source depleted by the extraction of the earlier melt phase. Some of the massive peridotites in the UMAA may either be cumulates or represent metasomatized and remobilized upper mantle. The Chilas Complex shows similarities with many other (supra)subduction‐related mafic–ultramafic complexes worldwide.     

Impact of draining hilly lands on runoff and on‐site erosion: a case study from humid Ethiopia

The use of drainage ditches on farmland has an impact on erosion processes both on‐site and off‐site, though their environmental impacts are not unequivocal. Here we study the runoff response and related rill erosion after installing drainage ditches and assess the effects of stone bunds in north Ethiopia. Three different land management systems were studied in 10 cropland catchments around Wanzaye during the rainy season of 2013: (1) the exclusive use of drainage ditches (locally called feses), (2) the exclusive use of stone bunds, and (3) a mixture of both systems. Stone bunds are an effective soil and water conservation technique, making the land more resistant against on‐site erosion, and allowing feses to be installed at a larger angle with the contour. The mean rill volumes for the 10 studied cropland catchments during the rainy season of 2013 was 3.73 ± 4.20 m³ ha^?1 corresponding to a soil loss of 5.72 ± 6.30 ton ha^?1. The establishment of feses causes larger rill volumes (R = 0.59, N = 10), although feses are perceived as the best way to avoid soil erosion when no stone bunds are present. The use of feses increases event‐based runoff coefficients (RCs) on cropland from c. 5% to values up to 39%. Also, a combination of low stone bund density and high feses density results in a higher RC, whereas catchments with a high stone bund density and low feses density have a lower RC. Peak runoff discharges decrease when stone bund density increases, whereas feses density is positively related to the peak runoff discharge. A multiple linear relation in which both feses and stone bund densities are used as explanatory variable, performs best in explaining runoff hydrograph peakedness (R² = 83%). Copyright © 2015 John Wiley & Sons, Ltd.     

The role of long‐term human impact on avulsion and fan development

This study aims to understand (mainly qualitatively) the long‐term role of human impact on avulsion processes and the development of fluvial (mega‐) fans in semi‐arid environments. In this paper we refer to human impact as the direct influences of actions on the river's hydraulics (i.e. flow regulation, flow diversion and channel engineering). In five case‐studies drawn from the Khuzestan plains in southwest Iran we have analysed the setup and triggering conditions of specific avulsions that occurred in the past (timescale of millennia) and identified the role of human interference in their causation. Our analysis is based on the integration of historical, archaeological, geomorphological and geological data. Through this study we demonstrate that avulsions in the Khuzestan plains are the result of long‐term and complex interplay between multiple human‐induced and natural causes. In similar ways human‐induced actions may play important roles during different phases of avulsion development. The ‘success‘ of an avulsion in the post‐triggering phase may be defined by human‐induced setup causes as well as morphodynamic processes. We suggest that present‐day flood events may be partly inherited from long‐term human alterations of the natural processes. These finding could have implications for any fluvial system (e.g. distributive fluvial systems, deltas) where avulsion plays a major role in their development and research tends to emphasize on natural mechanisms. Copyright © 2016 John Wiley & Sons, Ltd.     

Application of wavelet-artificial intelligence hybrid models for water quality prediction: a case study in Aji-Chay River,Iran

The accuracy of Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), wavelet-ANN and wavelet-ANFIS in predicting monthly water salinity levels of northwest Iran’s Aji-Chay River was assessed. The models were calibrated, validated and tested using different subsets of monthly records (October 1983 to September 2011) of individual solute (Ca²⁺, Mg²⁺, Na⁺, SO₄ ^2? and Cl^?) concentrations (input parameters, meq L^?1), and electrical conductivity-based salinity levels (output parameter, µS cm^?1), collected by the East Azarbaijan regional water authority. Based on the statistical criteria of coefficient of determination (R²), normalized root mean square error (NRMSE), Nash–Sutcliffe efficiency coefficient (NSC) and threshold statistics (TS) the ANFIS model was found to outperform the ANN model. To develop coupled wavelet-AI models, the original observed data series was decomposed into sub-time series using Daubechies, Symlet or Haar mother wavelets of different lengths (order), each implemented at three levels. To predict salinity input parameter series were used as input variables in different wavelet order/level-AI model combinations. Hybrid wavelet-ANFIS (R² = 0.9967, NRMSE = 2.9 × 10^?5 and NSC = 0.9951) and wavelet-ANN (R² = 0.996, NRMSE = 3.77 × 10^?5 and NSC = 0.9946) models implementing the db4 mother wavelet decomposition outperformed the ANFIS (R² = 0.9954, NRMSE = 3.77 × 10^?5 and NSC = 0.9914) and ANN (R² = 0.9936, NRMSE = 3.99 × 10^?5 and NSC = 0.9903) models.     

Recent marginal changes of the Mittivakkat Glacier,Southeast Greenland and the discovery of remains of reindeer (Rangifer tarandus), polar bear (Ursus maritimus) and peaty material

Geografisk Tidsskrift, Danish Journal of Geography 108(1):137–142, 2008

During field observations in August 2005 antler remains of a reindeer were found at a recently deglaciated site at about 500 m asl., and bones from a polar bear were found at about 300 m asl. along the margin of Mittivakkat Glacier, Southeast Greenland. Radio carbon dating determined the age of the samples to 720 ¹⁴C years and 350 ¹⁴C years, respectively. In August 2006 old surface vegetation covering peaty material became exposed due to ice recession close to the site where the antler was found. The radio carbon age of small roots from the material was determined to 1530 ¹⁴C years, and is in agreement with dating of woody remains of Salix glauca found close by, at the top of a nearby nunatak in 1999. The antler indicates that reindeer lived in the area when the glacier began to advance from a position where it was close to or smaller than today. The vegetation surface and peaty material indicate that the climate was warmer before the onset of the Little Ice Age in Southeast Greenland than today.