Search for: All records

Abstract Rock glaciers dominate the cryosphere in mid‐latitude alpine settings, yet their activity and their histories remain challenging to constrain. We focus on the Thomas Lake rock glacier on Mt. Sopris in Colorado, USA. We measure surface velocities by feature tracking of image pairs and document Holocene¹⁰Be exposure ages on surface debris. The surface speeds average 0.8 m/yr and peak at 2 m/yr in a steep reach. Exposure ages range from 1.4 to 13.2 kyr and monotonically increase down‐glaciers. Ages exceeding 6 kyr occur in the bottom quarter of the landform, coinciding with sporadic tree cover. These constraints constrain a numerical model of Holocene rock glacier activity. In our model, surface velocity is entirely explained by the deformation of the ice‐rich core with the extra load of the rocky carapace. Surface mass balance is simplified to an accumulation area of ice and debris equivalent to the avalanche cone, and very low, uniform ablation in the remaining rock glacier where rock cover minimizes melt. Climate drives the activity through a history of ice accumulation in the avalanche cone. Matching the observed age and speed structure requires: (a) Early Holocene growth of the rock glacier, (b) low accumulation during the middle Holocene warm period (Hypsithermal), and (c) two Neoglacial accumulation pulses, the most recent being the Little Ice Age. Pulses travel down the valley as kinematic waves, re‐activating the landform. The headwall retreat rate of 4 mm/yr, inferred from rocky layer thickness and surface speed, far outpaces bedrock down wearing rates. 

Increased access to high-resolution topography has revolutionized our ability to map out fine-scale topographic features at watershed to landscape scales. As our “vision” of the land surface has improved, so has the need for more robust quantification of the accuracy of the geomorphic maps we derive from these data. One broad class of mapping challenges is that of binary classification whereby remote sensing data are used to identify the presence or absence of a given feature. Fortunately, there is a large suite of metrics developed in the data sciences well suited to quantifying the pixel-level accuracy of binary classifiers. This analysis focuses on how these metrics perform when there is a need to quantify how the number and extent of landforms are expected to vary as a function of the environmental forcing (e.g., due to climate, ecology, material property, erosion rate). Results from a suite of synthetic surfaces show how the most widely used pixel-level accuracy metric, the F1 score, is particularly poorly suited to quantifying accuracy for this kind of application. Well-known biases to imbalanced data are exacerbated by methodological strategies that calibrate and validate classifiers across settings where feature abundances vary. The Matthews correlation coefficient largely removes this bias over a wide range of feature abundances such that the sensitivity of accuracy scores to geomorphic setting instead embeds information about the size and shape of features and the type of error. If error is random, the Matthews correlation coefficient is insensitive to feature size and shape, though preferential modification of the dominant class can limit the domain over which scores can be compared. If the error is systematic (e.g., due to co-registration error between remote sensing datasets), this metric shows strong sensitivity to feature size and shape such that smaller features with more complex boundaries induce more classification error. Future studies should build on this analysis by interrogating how pixel-level accuracy metrics respond to different kinds of feature distributions indicative of different types of surface processes. 

Abstract Large dendritic valley networks observed on Mars present a paleoclimate paradox. Geologic observations of Noachian units on Mars reveal a global extent of valley networks, which are believed to have been formed through incisions made by flowing water. However, most climate models predict global surface temperatures too far below the freezing point of water to support an active hydrological system. Conflicting observations and models have led to disparate theories for the climate of early Mars. In this work, we surveyed a large region of the cratered southern highlands to identify the location, elevation, and distribution of observed valley heads. These valley head locations were compared to landscape evolution simulations in which the spatial distribution of runoff was varied. The measured valley head distributions were compared to predictions from landscape evolution models for two end‐member hypotheses: (a) a warm wet climate that supported spatially distributed precipitation, and (b) surface runoff from ice cap margins, as envisioned by the Late Noachian Icy Highland model (LNIH). The observed elevation distribution in valley heads is consistent with the prediction of precipitation‐fed models, and inconsistent with models in which runoff derives exclusively from a single line‐source of high‐elevation ice‐melt. The results support the view that it is unlikely for ice caps to be the sole source of water and are consistent with the hypothesis that precipitation significantly contributed to valley network formation on ancient Mars. 

Effects of a changing climate on agricultural system productivity are poorly understood, and likely to be met with as yet undefined agricultural adaptations by farmers and associated business and governmental entities. The continued vitality of agricultural systems depends on economic conditions that support farmers’ livelihoods. Exploring the long-term effects of adaptations requires modeling agricultural and economic conditions to engage stakeholders upon whom the burden of any adaptation will rest. Here, we use a new freeware model FEWCalc (Food-Energy-Water Calculator) to project farm incomes based on climate, crop selection, irrigation practices, water availability, and economic adaptation of adding renewable energy production. Thus, FEWCalc addresses United Nations Global Sustainability Goals No Hunger and Affordable and Clean Energy. Here, future climate scenario impacts on crop production and farm incomes are simulated when current agricultural practices continue so that no agricultural adaptations are enabled. The model Decision Support System for Agrotechnology Transfer (DSSAT) with added arid-region dynamics is used to simulate agricultural dynamics. Demonstrations at a site in the midwest USA with 2008–2017 historical data and two 2018–2098 RCP climate scenarios provide an initial quantification of increased agricultural challenges under climate change, such as reduced crop yields and increased financial losses. Results show how this finding is largely driven by increasing temperatures and changed distribution of precipitation throughout the year. Without effective technological advances and operational and policy changes, the simulations show how rural areas could increasingly depend economically on local renewable energy, while agricultural production from arid regions declines by 50% or more. 

Abstract Hillslope topographic change in response to climate and climate change is a key aspect of landscape evolution. The impact of short‐duration rainstorms on hillslope evolution in arid regions is persistently questioned but often not directly examined in landscape evolution studies, which are commonly based on mean climate proxies. This study focuses on hillslope surface processes responding to rainstorms in the driest regions of Earth. We present a numerical model for arid, rocky hillslopes with lithology of a softer rock layer capped by a cliff‐forming resistant layer. By representing the combined action of bedrock and clast weathering, cliff‐debris ravel, and runoff‐driven erosion, the model can reproduce commonly observed cliff‐profile morphology. Numerical experiments with a fixed base level were used to test hillslope response to cliff‐debris grain size, rainstorm intensities, and alternation between rainstorm patterns. The persistence of vertical cliffs and the pattern of sediment sorting depend on rainstorm intensities and the size of cliff debris. Numerical experiments confirm that these two variables could have driven the landscape in the Negev Desert (Israel) toward an observed spatial contrast in topographic form over the past 10⁵–10⁶ years. For a given total storm rain depth, short‐duration higher‐intensity rainstorms are more erosive, resulting in greater cliff retreat distances relative to longer, low‐intensity storms. Temporal alternation between rainstorm regimes produces hillslope profiles similar to those previously attributed to Quaternary oscillations in the mean climate. We suggest that arid hillslopes may undergo considerable geomorphic transitions solely by alternating intra‐storm patterns regardless of rainfall amounts. 

Abstract The impact of climate on topography, which is a theme in landscape evolution studies, has been demonstrated, mostly, at mountain range scales and across climate zones. However, in drylands, spatiotemporal discontinuities of rainfall and the crucial role of extreme rainstorms raise questions and challenges in identifying climate properties that govern surface processes. Here, we combine methods to examine hyperarid escarpment sensitivity to storm‐scale forcing. Using a high‐resolution DEM and field measurements, we analyzed the topography of a 40‐km‐long escarpment in the Negev desert (Israel). We also used rainfall intensity data from a convection‐permitting numerical weather model for storm‐scale statistical analysis. We conducted hydrological simulations of synthetic rainstorms, revealing the frequency of sediment mobilization along the sub‐cliff slopes. Results show that cliff gradients along the hyperarid escarpment increase systematically from the wetter (90 mm yr⁻¹) southwestern to the drier (45 mm yr⁻¹) northeastern sides. Also, sub‐cliff slopes at the southwestern study site are longer and associated with milder gradients and coarser sediments. Storm‐scale statistical analysis reveals a trend of increasing extreme (>10 years return‐period) intensities toward the northeast site, opposite to the trend in mean annual rainfall. Hydrological simulations based on these statistics indicate a higher frequency of sediment mobilization in the northeast, which can explain the pronounced topographic differences between the sites. The variations in landscape and rainstorm properties across a relatively short distance highlight the sensitivity of arid landforms to extreme events. 

Abstract Generalizable relationships for how subdaily rainfall statistics imprint into runoff statistics are lacking. We use the Colorado Front Range, known for destructive rainfall‐triggered floods and landslides, to assess whether orographic patterns in runoff generation are a direct consequence of rainstorm climatology. Climatological analysis relies on a dense network of tipping‐bucket rain gauges and gridded precipitation frequency estimates from the National Oceanic and Atmospheric Administration to evaluate relationships among subdaily rainfall statistics, topography, and flood frequency throughout the South Platte River basin. We find that event‐scale rainfall statistics only weakly depend on elevation, suggesting that orographic gradients in runoff “extremes” are not simply a consequence of rainfall patterns. In contrast, bedrock exposure strongly varies with elevation in a way that plausibly explains enhanced runoff generation at lower elevations via reduced water storage capacity. These findings are suggestive of feedbacks between bedrock river evolution and hillslope hydrology not typically included in models of landscape evolution.