Deep learning to extract the meteorological by-catch of wildlife cameras.

Microclimate-proximal climatic variation at scales of metres and minutes-can exacerbate or mitigate the impacts of climate change on biodiversity. However, most microclimate studies are temperature centric, and do not consider meteorological factors such as sunshine, hail and snow. Meanwhile, remote cameras have become a primary tool to monitor wild plants and animals, even at micro-scales, and deep learning tools rapidly convert images into ecological data. However, deep learning applications for wildlife imagery have focused exclusively on living subjects. Here, we identify an overlooked opportunity to extract latent, ecologically relevant meteorological information. We produce an annotated image dataset of micrometeorological conditions across 49 wildlife cameras in South Africa's Maloti-Drakensberg and the Swiss Alps. We train ensemble deep learning models to classify conditions as overcast, sunshine, hail or snow. We achieve 91.7% accuracy on test cameras not seen during training. Furthermore, we show how effective accuracy is raised to 96% by disregarding 14.1% of classifications where ensemble member models did not reach a consensus. For two-class weather classification (overcast vs. sunshine) in a novel location in Svalbard, Norway, we achieve 79.3% accuracy (93.9% consensus accuracy), outperforming a benchmark model from the computer vision literature (75.5% accuracy). Our model rapidly classifies sunshine, snow and hail in almost 2 million unlabelled images. Resulting micrometeorological data illustrated common seasonal patterns of summer hailstorms and autumn snowfalls across mountains in the northern and southern hemispheres. However, daily patterns of sunshine and shade diverged between sites, impacting daily temperature cycles. Crucially, we leverage micrometeorological data to demonstrate that (1) experimental warming using open-top chambers shortens early snow events in autumn, and (2) image-derived sunshine marginally outperforms sensor-derived temperature when predicting bumblebee foraging. These methods generate novel micrometeorological variables in synchrony with biological recordings, enabling new insights from an increasingly global network of wildlife cameras.

Contamination Pattern and Risk Assessment of Polar Compounds in Snow Melt: An Integrative Proxy of Road Runoffs.

To assess the contamination and potential risk of snow melt with polar compounds, road and background snow was sampled during a melting event at 23 sites at the city of Leipzig and screened for 489 chemicals using liquid chromatography high-resolution mass spectrometry with target screening. Additionally, six 24 h composite samples were taken from the influent and effluent of the Leipzig wastewater treatment plant (WWTP) during the snow melt event. 207 compounds were at least detected once (concentrations between 0.80 ng/L and 75 µg/L). Consistent patterns of traffic-related compounds dominated the chemical profile (58 compounds in concentrations from 1.3 ng/L to 75 µg/L) and among them were 2-benzothiazole sulfonic acid and 1-cyclohexyl-3-phenylurea from tire wear and denatonium used as a bittern in vehicle fluids. Besides, the analysis unveiled the presence of the rubber additive 6-PPD and its transformation product N-(1.3-dimethylbutyl)-N'-phenyl-p-phenylenediamine quinone (6-PPDQ) at concentrations known to cause acute toxicity in sensitive fish species. The analysis also detected 149 other compounds such as food additives, pharmaceuticals, and pesticides. Several biocides were identified as major risk contributors, with a more site-specific occurrence, to acute toxic risks to algae (five samples) and invertebrates (six samples). Ametryn, flumioxazin, and 1,2-cyclohexane dicarboxylic acid diisononyl ester are the main compounds contributing to toxic risk for algae, while etofenprox and bendiocarb are found as the main contributors for crustacean risk. Correlations between concentrations in the WWTP influent and flow rate allowed us to discriminate compounds with snow melt and urban runoff as major sources from other compounds with other dominant sources. Removal rates in the WWTP showed that some traffic-related compounds were largely eliminated (removal rate higher than 80%) during wastewater treatment and among them was 6-PPDQ, while others persisted in the WWTP.

Earlier snow melt and reduced summer precipitation alter floral traits important to pollination.

Climate change can cause changes in expression of organismal traits that influence fitness. In flowering plants, floral traits can respond to drought, and that phenotypic plasticity has the potential to affect pollination and plant reproductive success. Global climate change is leading to earlier snow melt in snow-dominated ecosystems as well as affecting precipitation during the growing season, but the effects of snow melt timing on floral morphology and rewards remain unknown. We conducted crossed manipulations of spring snow melt timing (early vs. control) and summer monsoon precipitation (addition, control, and reduction) that mimicked recent natural variation, and examined plastic responses in floral traits of Ipomopsis aggregata over 3 years in the Rocky Mountains. We tested whether increased summer precipitation compensated for earlier snow melt, and if plasticity was associated with changes in soil moisture and/or leaf gas exchange. Lower summer precipitation decreased corolla length, style length, corolla width, sepal width, and nectar production, and increased nectar concentration. Earlier snow melt (taking into account natural and experimental variation) had the same effects on those traits and decreased inflorescence height. The effect of reduced summer precipitation was stronger in earlier snow melt years for corolla length and sepal width. Trait reductions were explained by drier soil during the flowering period, but this effect was only partially explained by how drier soils affected plant water stress, as measured by leaf gas exchange. We predicted the effects of plastic trait changes on pollinator visitation rates, pollination success, and seed production using prior studies on I. aggregata. The largest predicted effect of drier soil on relative fitness components via plasticity was a decrease in male fitness caused by reduced pollinator rewards (nectar production). Early snow melt and reduced precipitation are strong drivers of phenotypic plasticity, and both should be considered when predicting effects of climate change on plant traits in snow-dominated ecosystems.

Ecological responses to variation in seasonal snow cover.

Seasonal snow is among the most important factors governing the ecology of many terrestrial ecosystems, but rising global temperatures are changing snow regimes and driving widespread declines in the depth and duration of snow cover. Loss of the insulating snow layer will fundamentally change the environment. Understanding how individuals, populations, and communities respond to different snow conditions is thus essential for predicting and managing future ecosystem change. We synthesized 365 studies that examined ecological responses to variation in winter snow conditions. This research encompasses a broad range of methods (experimental manipulations, measurement of natural snow gradients, and long-term monitoring), locations (35 countries), study organisms (plants, mammals, arthropods, birds, fish, lichen, and fungi), and response measures. Earlier snowmelt was consistently associated with advanced spring phenology in plants, mammals, and arthropods. Reduced snow depth often increased mortality or physical injury in plants, although there were few clear effects on animals. Neither snow depth nor snowmelt timing had clear or consistent directional effects on body size of animals or biomass of plants. However, because 96% of studies were from the northern hemisphere, the generality of these trends across ecosystems and localities is also unclear. We identified substantial research gaps for several taxonomic groups and response types; research on wintertime responses was notably scarce. Future research should prioritize examination of the mechanisms underlying responses to changing snow conditions and the consequences of those responses for seasonally snow-covered ecosystems.

Respuestas Ecológicas a la Variación de la Cobertura Estacional de Nieve Resumen La nieve estacional se encuentra entre los factores más importantes que determinan la ecología de muchos ecosistemas terrestres, pero las crecientes temperaturas mundiales están cambiando los sistemas de nieve y causando declinaciones generalizadas en la profundidad y la duración de la capa de nieve. La pérdida de la capa de nieve aislante cambiará fundamentalmente el ambiente. El entendimiento de cómo los individuos, las poblaciones y las comunidades responden a las diferentes condiciones de nieve es esencial para predecir y manejar los cambios del ecosistema en el futuro. Sintetizamos 365 estudios que examinaron las respuestas ecológicas a la variación en las condiciones invernales de nieve. Esta investigación engloba una gama amplia de métodos (manipulaciones experimentales, medida de los gradientes naturales de nieve y monitoreo a largo plazo), localidades (35 países), organismos de estudio (plantas, mamíferos, artrópodos, aves, peces, líquenes y hongos) y medidas de respuesta. El deshielo temprano estuvo asociado continuamente con el adelanto de la fenología de plantas, mamíferos y artrópodos en primavera. La reducción de la profundidad de la nieve con frecuencia incrementó la mortalidad o las lesiones físicas en las plantas, aunque tuvo pocos efectos visibles sobre los animales. Ni la profundidad de la nieve ni la temporalidad del deshielo tuvieron efectos direccionales claros o consistentes sobre el tamaño corporal de los animales o la biomasa de las plantas. Sin embargo, ya que el 96% de los estudios se realizó en el hemisferio norte, la generalidad de estas tendencias en todos los ecosistemas y localidades tampoco está clara. Identificamos vacíos importantes en la investigación en torno a varios grupos taxonómicos y los tipos de respuesta; la información sobre las respuestas invernales estaba particularmente reducida. Las futuras investigaciones deberían priorizar el análisis de los mecanismos subyacentes a las respuestas ante las condiciones cambiantes de nieve y las consecuencias de aquellas respuestas para los ecosistemas cubiertos de nieve.

Snow melt stimulates ecosystem respiration in Arctic ecosystems.

Cold seasons in Arctic ecosystems are increasingly important to the annual carbon balance of these vulnerable ecosystems. Arctic winters are largely harsh and inaccessible leading historic data gaps during that time. Until recently, cold seasons have been assumed to have negligible impacts on the annual carbon balance but as data coverage increases and the Arctic warms, the cold season has been shown to account for over half of annual methane (CH4 ) emissions and can offset summer photosynthetic carbon dioxide (CO2 ) uptake. Freeze-thaw cycle dynamics play a critical role in controlling cold season CO2 and CH4 loss, but the relationship has not been extensively studied. Here, we analyze freeze-thaw processes through in situ CO2 and CH4 fluxes in conjunction with soil cores for physical structure and porewater samples for redox biogeochemistry. We find a movement of water toward freezing fronts in soil cores, leaving air spaces in soils, which allows for rapid infiltration of oxygen-rich snow melt in spring as shown by oxidized iron in porewater. The snow melt period coincides with rising ecosystem respiration and can offset up to 41% of the summer CO2 uptake. Our study highlights this important seasonal process and shows spring greenhouse gas emissions are largely due to production from respiration instead of only bursts of stored gases. Further warming is projected to result in increases of snowpack and deeper thaws, which could increase this ecosystem respiration dominate snow melt period causing larger greenhouse gas losses during spring.

Local snow melt and temperature-but not regional sea ice-explain variation in spring phenology in coastal Arctic tundra.

The Arctic is undergoing dramatic environmental change with rapidly rising surface temperatures, accelerating sea ice decline and changing snow regimes, all of which influence tundra plant phenology. Despite these changes, no globally consistent direction of trends in spring phenology has been reported across the Arctic. While spring has advanced at some sites, spring has delayed or not changed at other sites, highlighting substantial unexplained variation. Here, we test the relative importance of local temperatures, local snow melt date and regional spring drop in sea ice extent as controls of variation in spring phenology across different sites and species. Trends in long-term time series of spring leaf-out and flowering (average span: 18 years) were highly variable for the 14 tundra species monitored at our four study sites on the Arctic coasts of Alaska, Canada and Greenland, ranging from advances of 10.06 days per decade to delays of 1.67 days per decade. Spring temperatures and the day of spring drop in sea ice extent advanced at all sites (average 1°C per decade and 21 days per decade, respectively), but only those sites with advances in snow melt (average 5 days advance per decade) also had advancing phenology. Variation in spring plant phenology was best explained by snow melt date (mean effect: 0.45 days advance in phenology per day advance snow melt) and, to a lesser extent, by mean spring temperature (mean effect: 2.39 days advance in phenology per °C). In contrast to previous studies examining sea ice and phenology at different spatial scales, regional spring drop in sea ice extent did not predict spring phenology for any species or site in our analysis. Our findings highlight that tundra vegetation responses to global change are more complex than a direct response to warming and emphasize the importance of snow melt as a local driver of tundra spring phenology.

Plasticity in functional traits in the context of climate change: a case study of the subalpine forb Boechera stricta (Brassicaceae).

Environmental variation often induces shifts in functional traits, yet we know little about whether plasticity will reduce extinction risks under climate change. As climate change proceeds, phenotypic plasticity could enable species with limited dispersal capacity to persist in situ, and migrating populations of other species to establish in new sites at higher elevations or latitudes. Alternatively, climate change could induce maladaptive plasticity, reducing fitness, and potentially stalling adaptation and migration. Here, we quantified plasticity in life history, foliar morphology, and ecophysiology in Boechera stricta (Brassicaceae), a perennial forb native to the Rocky Mountains. In this region, warming winters are reducing snowpack and warming springs are advancing the timing of snow melt. We hypothesized that traits that were historically advantageous in hot and dry, low-elevation locations will be favored at higher elevation sites due to climate change. To test this hypothesis, we quantified trait variation in natural populations across an elevational gradient. We then estimated plasticity and genetic variation in common gardens at two elevations. Finally, we tested whether climatic manipulations induce plasticity, with the prediction that plants exposed to early snow removal would resemble individuals from lower elevation populations. In natural populations, foliar morphology and ecophysiology varied with elevation in the predicted directions. In the common gardens, trait plasticity was generally concordant with phenotypic clines from the natural populations. Experimental snow removal advanced flowering phenology by 7 days, which is similar in magnitude to flowering time shifts over 2-3 decades of climate change. Therefore, snow manipulations in this system can be used to predict eco-evolutionary responses to global change. Snow removal also altered foliar morphology, but in unexpected ways. Extensive plasticity could buffer against immediate fitness declines due to changing climates.

A Novel Method to Quantify Near-Surface Boundary-Layer Dynamics at Ultra-High Spatio-Temporal Resolution.

The lateral transport of heat above abrupt (sub-)metre-scale steps in land surface temperature influences the local surface energy balance. We present a novel experimental method to investigate the stratification and dynamics of the near-surface atmospheric layer over a heterogeneous land surface. Using a high-resolution thermal infrared camera pointing at synthetic screens, a 30 Hz sequence of frames is recorded. The screens are deployed upright and horizontally aligned with the prevailing wind direction. The screen's surface temperature serves as a proxy for the local air temperature. We developed a method to estimate near-surface two-dimensional wind fields at centimetre resolution from tracking the air temperature pattern on the screens. Wind field estimations are validated with near-surface three-dimensional short-path ultrasonic data. To demonstrate the capabilities of the screen method, we present results from a comprehensive field campaign at an alpine research site during patchy snow cover conditions. The measurements reveal an extremely heterogeneous near-surface atmospheric layer. Vertical profiles of horizontal and vertical wind reflect multiple layers of different static stability within 2 m above the surface. A dynamic, thin stable internal boundary layer (SIBL) develops above the leading edge of snow patches protecting the snow surface from warmer air above. During pronounced gusts, the warm air from aloft entrains into the SIBL and reaches down to the snow surface adding energy to the snow pack. Measured vertical turbulent sensible heat fluxes are shown to be consistent with air temperature and wind profiles obtained using the screen method and confirm its capabilities to investigate complex in situ near-surface heat exchange processes. Supplementary Information: The online version contains supplementary material available at 10.1007/s10546-022-00752-3.

Influence of aerosol radiative effects on surface temperature and snow melt in the Himalayan region.

Recent increases in surface temperature and snow melt acceleration in the Himalayan region are influenced by many factors. Here we investigate the influence of absorbing aerosols, including black carbon and dust, on surface temperature and snow melt in western, central, and eastern parts of the India-Nepal Himalayan region (INHR). We compare 40-y simulations (1971-2010) one with all evolving forcing agents representative of a present-day aerosol scenario, compared to a low aerosol forcing scenario. The difference between these scenarios shows a significant increase in surface air temperature, with higher warming in parts of Western and Central Himalaya (~0.2-2 °C) in the months of April and May. Higher absorbing aerosol (BC and dust abundance) both at the surface and in the atmospheric column, in the present-day aerosol simulations, led to increases in atmospheric radiative forcing and surface shortwave heating rate forcing (SWHRF), compared to the low aerosol forcing case. Therefore, the absorbing aerosols cause anomalous atmospheric heat energy transfer to land due to high surface SWHRF and changes in surface energy flux, leading to snow melt. The present model version did not parameterize snow albedo feedback, which would increase the magnitudes of the changes simulated here.

Domino effect of a natural cascade alpine lake system on the Third Pole.

Third Pole natural cascade alpine lakes (NCALs) are exceptionally sensitive to climate change, yet the underlying cryosphere-hydrological processes and associated societal impacts are largely unknown. Here, with a state-of-the-art cryosphere-hydrology-lake-dam model, we quantified the notable high-mountain Hoh-Xil NCALs basin (including Lakes Zonag, Kusai, Hedin Noel, and Yanhu, from upstream to downstream) formed by the Lake Zonag outburst in September 2011. We demonstrate that long-term increased precipitation and accelerated ice and snow melting as well as short-term heavy precipitation and earthquake events were responsible for the Lake Zonag outburst; while the permafrost degradation only had a marginal impact on the lake inflows but was crucial to lakeshore stability. The quadrupling of the Lake Yanhu area since 2012 was due to the tripling of inflows (from 0.25 to 0.76 km3/year for 1999 to 2010 and 2012 to 2018, respectively). Prediction of the NCALs changes suggests a high risk of the downstream Qinghai-Tibet Railway, necessitating timely adaptions/mitigations.

Best environmental predictors of breeding phenology differ with elevation in a common woodland bird species.

Temperatures in mountain areas are increasing at a higher rate than the Northern Hemisphere land average, but how fauna may respond, in particular in terms of phenology, remains poorly understood. The aim of this study was to assess how elevation could modify the relationships between climate variability (air temperature and snow melt-out date), the timing of plant phenology and egg-laying date of the coal tit (Periparus ater). We collected 9 years (2011-2019) of data on egg-laying date, spring air temperature, snow melt-out date, and larch budburst date at two elevations (~1,300 m and ~1,900 m asl) on a slope located in the Mont-Blanc Massif in the French Alps. We found that at low elevation, larch budburst date had a direct influence on egg-laying date, while at high-altitude snow melt-out date was the limiting factor. At both elevations, air temperature had a similar effect on egg-laying date, but was a poorer predictor than larch budburst or snowmelt date. Our results shed light on proximate drivers of breeding phenology responses to interannual climate variability in mountain areas and suggest that factors directly influencing species phenology vary at different elevations. Predicting the future responses of species in a climate change context will require testing the transferability of models and accounting for nonstationary relationships between environmental predictors and the timing of phenological events.

Water-soluble carbohydrates in Patzkea paniculata (L.): a plant strategy to tolerate snowpack reduction and spring drought in subalpine grasslands.

In subalpine grasslands of the central French Alps, cessation of traditional mowing promotes dominance of Patzkea paniculata (L.) G.H.Loos (Poaceae) tussocks, with high biomass but low fodder quality. Mowing limits P. paniculata abundance through the depletion of its water-soluble carbohydrate (WSC) reserves, which sustain early spring growth initiation. However, the effectiveness of mowing effects is modulated by grassland functional composition, fertilization and climate change, as WSC compounds, and notably fructans, support plant physiological responses to climate stresses such as drought or frost. To characterize the mechanisms underpinning the control of P. paniculata under global change, we tested the effects of climate manipulation (combined snow removal and drought) and management (cutting and fertilization) alone or in combination on P. paniculata WSC storage in assembled grassland communities of varying functional composition. Management and climate treatments individually decreased seasonal fructan storage, with neither additive nor synergic effects between them, primarily due to the dominance of management over climate effects. Fructan amounts were higher in individuals growing in unmanaged exploitative communities compared to unmanaged conservative communities, regardless of climate treatments, but management overrode these differences. Our findings suggest that reduction by combined snow removal and drought of P. paniculata carbon allocation to WSC storage may similarly limit its dominance to that in current mowing practices.

Light-absorbing impurities in snow of the Indian Western Himalayas: impact on snow albedo, radiative forcing, and enhanced melting.

Seasonal snow cover in the Himalayas acts as source of fresh water for several Asian rivers such as Indus, Ganges, Brahmaputra, and Yangtze. Early loss of seasonal snow exposes the ice layer of the glaciers directly to sunlight, consequently leading to their ablation and alterations in discharge of glacier-fed rivers. Therefore, any alteration in the melting rate of the Himalayan snow pack can significantly affect the ecological balance in the region. Besides global warming, enhanced melting of snow, caused by light-absorbing impurities (LAIs) such as dust and elemental carbon (EC), has also been recognized as prominent cause of enhanced melting of snow in the Himalayas of China and Nepal. However, in light of vast area of the Himalayas and persistent emissions from India, studies, emphasizing the potential of LAIs to substantially affect the snow radiation budget of snow cover in IWHs, are still scanty. Therefore, in this study, field campaigns were made on three glaciers, i.e., Hamta, Beas Kund, and Deo Tibba, in IWHs to collect snow samples for estimation of LAIs. Snow of the studied glaciers was observed to be contaminated with 13.02 to 74.57 ng/g of EC and 32.14 to 216.54 µg/g of dust. Albedo simulations done using SNow and ICe Aerosol Radiation (SNICAR) model indicated that besides the changes caused by increased grain size, EC and dust, cumulatively induced 0.60 to 32.65% reduction in albedo of snow. Further assessment, constrained by measurements, illustrated that radiative forcing (RF), of 1.8 to 80 W/m2, was instigated due to enhanced thermal absorption of snow. Ten hours of daily mean RFs in this range could correspond to 3 to 9.65 mm/d of snow melt and contribute significantly in reducing the seasonal snow cover in IWHs. Considering the consequences of LAIs-induced snow melt and lack of in situ observations in the IWHs, the outcomes of this study could assist researchers and policy makers in developing efficient climate models and framing mitigation measures, respectively.

Citizen science evidence from the past century shows that Scottish rivers are warming.

Salmonid species are highly sensitive to river water temperature. Although long-term river temperature monitoring is essential for assessing drivers of change in ecological systems, these data are rarely available from statutory monitoring. We utilized a 105-year citizen science data set of river water temperature from the River Spey, North-East Scotland, gathered during the fishing season (April-October) between 1912 and 2016. As there were gaps in the records we applied generalised additive models to reconstruct long-term daily river temperature in the fishing season from air temperature, cumulative air temperature, day length and runoff. For that, continuous hydrometeorological data have been obtained from statutory monitoring and process-based models. Long-term warming trends of river temperature, namely an increase of 0.2 K per decade after 1961, have been mostly related to increasing air temperature of the same magnitude. Indirect impacts of rising air temperatures include less snow accumulation and snow melt as well as earlier snow melt. The snow free period starts around 2 days earlier per decade throughout the study period and 7 days earlier per decade after 1965. Consequently, the contribution of snow melt and its cooling properties to river temperature in spring are declining. Citizen science delivered a data set that filled a vital knowledge gap in the long-term historical assessment of river temperatures. Such information provides a robust basis for future assessments of global change and can help inform decision-makers about the potential importance of enhancing the resilience of rivers and aquatic ecology to climate change.

Arctic Geese Tune Migration to a Warming Climate but Still Suffer from a Phenological Mismatch.

Climate warming challenges animals to advance their timing of reproduction [1], but many animals appear to be unable to advance at the same rate as their food species [2, 3]. As a result, mismatches can arise between the moment of largest food requirements for their offspring and peak food availability [4-6], with important fitness consequences [7]. For long-distance migrants, adjustment of phenology to climate warming may be hampered by their inability to predict the optimal timing of arrival at the breeding grounds from their wintering grounds [8]. Arrival can be advanced if birds accelerate migration by reducing time on stopover sites [9, 10], but a recent study suggests that most long-distance migrants are on too tight a schedule to do so [11]. This may be different for capital-breeding migrants, which use stopovers not only to fuel migration but also to acquire body stores needed for reproduction [12-14]. By combining multiple years of tracking and reproduction data, we show that a long-distance migratory bird (the barnacle goose, Branta leucopsis) accelerates its 3,000 km spring migration to advance arrival on its rapidly warming Arctic breeding grounds. As egg laying has advanced much less than arrival, they still encounter a phenological mismatch that reduces offspring survival. A shift toward using more local resources for reproduction suggests that geese first need to refuel body stores at the breeding grounds after accelerated migration. Although flexibility in body store use allows migrants to accelerate migration, this cannot solve the time constraint they are facing under climate warming.