Mosquitoes in recent years have increased greatly in numbers due to the rapidly changing climate and rising temperatures. With this change comes suitable habitats for mosquitoes which are the most efficient killers in all of the animal kingdom due to the number of death from mosquito-borne diseases. If we were able to pinpoint the certain areas that mosquitoes are most attracted to we could in theory slow or even prevent the spread of mosquitoes. In our project, the research was conducted to find a correlation between the color and size of the traps to the amount of mosquitos that are present. With our findings, we were able to conclude that the bigger the traps, the faster the mosquitoes would be attracted to that area. We also found that the different traps would hold similar densities of mosquito larvae per square inch.

Deep learning techniques are being increasingly used in earth science applications - from climate change modelling to feature extraction from remote sensing imagery, given their advantage of increased contextual and hierarchical feature representation. However, deep learning comes at an expense of extensive computational resources and long training time to achieve benchmark results. This study suggests time-optimized deep learning techniques for training deep convolutional networks for one of the most sought after feature extraction subsets – building extraction from satellite/aerial imagery. Building extraction is one of the most important tasks in the dynamic pipeline of urban applications such as urban planning and management, disaster management, urban mapping etc. among other geospatial applications. Automatically extracting buildings from remotely sensed imagery has always been a challenging task, given the spectral homogeneity of buildings with the non-building features as well as the complex structural diversity within the image. With the availability of high resolution open-source satellite and UAV data, deep learning techniques have greatly improved building extraction. However, training on such high resolution data requires the networks to be significantly deeper, resulting in long model training and inference times. This study proposes a combination of two time efficient methods to train a Dynamic Res-U-Net for building extraction in less time without decreasing the training parameters: 1) Using Cyclical Learning and SuperConvergence concepts by dynamically changing the learning rate while training the network to achieve very high accuracy in very less time and 2) Using a specific order to train the layers of the network(s) to specially have the last layers of the networks perform better, leading to an overall improved network performance in lesser time. Building extraction results are gauged using the metrics of Accuracy, Dice Score and Intersection over Union (IoU) and F1-Score. The metrics comparison of training the Res-U-Net in the conventional way vs the proposed techniques shows an evident optimisation in terms of time. Better results are achieved in lesser training epochs using the proposed time-optimised training techniques.

The redox potential of soils is critical in understanding the structure and function of ecosystems. Soil redox state strongly governs the speciation, bioavailability, and solubility of limiting nutrients like nitrogen and phosphorous in soil. It also drives the reactivity, mobility, and toxicity of redox-sensitive elements such as anthropogenic contaminants, affecting soil and groundwater quality by altering or retaining undesirable metals. Although these factors are highly variable among different landcovers and soil depths, limited studies try to link the redox-sensitive elements with soil physical and chemical properties in various depths and with different landcovers. With designed experiments in Brazos River corridor in Texas, we (1) evaluate the effect of different land use and land covers on the concentrations of electron acceptors (O2, NO3, and SO4), reduced products (Mn(II), Fe(II)), and C, N, and P pools in the surface and deep soils, (2) determine effects of climatic gradient on redox biogeochemistry in deep soils, and (3) investigate the effects of soil physical and hydraulic properties on redox biogeochemistry. Soil physical and chemical properties were determined from varying soil depths (land surface up to 15 m) in different landcovers (grassland, forest, and salt marsh). Higher carbon and nitrogen content were observed in the surface soils due to carbon mineralization in all land covers. However, the phosphorous content was higher at 15-30 cm soil depth, because of the co-existing high iron and aluminum oxides concentrations that provide high surface area for phosphorous adsorption. C, N, P, and other redox-sensitive elements were positively correlated to clay content at various depths. The biogeochemical properties, including ammonium, ferric iron and sulfate, were disproportionally higher at the interface of soil layers where soil texture and hydraulic properties change. This finding reflects the role of soil layers as hot spots of biogeochemical processes in the subsurface. With the climate gradient across the study river basin, our data indicates C, N, P, and other redox-sensitive elements are more profound in the salt marsh and forest covers with higher annual mean temperature and precipitation as these factors stimulate microbial activity and thus influence redox processes.

Nearly one million Rohingya refugees of Myanmar have fled to neighboring Bangladesh for seeking shelter from systematic oppression since 25 August 2017. The speed and scale of the influx has led to unprecedented growth of the Kutupalong refugee camp of eastern coast of Bangladesh within few months. There is a few study focused on environmental degradation of the refugee camp, however, no study has been done so far to estimate the impact of camp expansion on ecosystem services loss. This study, therefore, made an attempt to estimate the changes of Ecosystem Services Value (ESV) in response to camp expansion for the year July 2017 (pre-camp) and July 2018 by GIS technique and corresponding global value coefficient developed by Costanza et.al. (1997). Land cover map of the study area was prepared by using Landsat 8 satellite data. Results show an overall decrease of vegetation of 2486 hectares, of which 20% were used to expand the camp and 80% were deforested. Total ecosystem service values of the study area reduced dramatically, from 51.53×106 US$ to 49.12×106 US$ in the study period. This represents a 4.63% net decline in annual value of ecosystem services in the study area. In terms of 2 km buffer of the camp, the net decline rate is found 32.58%. The significant changes are also recorded in individual ecosystem services function of the area. Hence, the findings of this study may motivate the Bangladesh government to develop better plans to protect the ecologically sensitive forested land and wildlife habitats surrounding the refugee camps and assist in more sustainable resource mobilization for the Rohingya refugees.

The Eastern Pontides are the northmost component of the Anatolian orogen. Its geological development closely associated with the evolution of the Ankara-Erzincan Suture. It exhibits records of the events from the opening to the eliminations of the surrounding oceans. During the Late Paleozoic, the Pontides were located in the north of Gondwana, facing the Paleo-Tethys Ocean. The southward subduction of the Paleo Tethyan oceanic lithosphere generated an active continental margin and opening of the Neo-Tethys Ocean as a back-arc basin during the Early Mesozoic. Throughout the Jurassic-Early Cretaceous, the Pontides remained a passive continental margin facing the Neo-Tethys in the south. Arc reversal occurred as the Neo-Tethys began subducting under the Pontides during the late Early Cretaceous (?)-Late Cretaceous. The Pontides experienced four collisional events throughout the development of the Ankara-Erzincan Suture; (1)- a forearc-arc collision occurred when the accretionary complex, which formed along the southern edge of the Pontides was backthrust over its leading edge during the Late Campanian. (2)- This was followed by a continent-arc collision when the Kırşehir Massif and the underlying NeoTethyan ophiolite nappe collided with and thrust over the Pontides at the end of the Early Eocene, (3)- Following the oceanic lithosphere’s total demise, the remnant basin located between the Pontides, and the Taurus was closed under the northerly advancing Taurus nappes during the Late Eocene. The latest collision is related with the collision of the Arabian Plate with the Anatolian plates. The Arabian Plate’s continuing northward advance after the demise of the NeoTethyan Ocean squeezed and shortened the Eastern Anatolia. From this time onward, the Eastern Pontides were thrust to the north and the south over the surrounding tectonic belts and started to rise as a coherent block.

The CNRM-Cerfacs Climate Prediction System (C3PS) is a new research modeling tool for performing climate reanalyses and seasonal-to-multiannual predictions for a wide array of earth system variables. C3PS is based on the CNRM-ESM2-1 model including interactive aerosols and stratospheric chemistry schemes as well as terrestrial and marine biogeochemistry enabling a comprehensive representation of the global carbon cycle. C3PS operates through a seamless coupled initialization for the atmosphere, land, ocean, sea ice and biogeochemistry components that allows a continuum of predictions across seasonal to interannual time-scales. C3PS has also contributed to the Decadal Climate Prediction Project (DCPP-A) as part of the sixth Coupled Model Intercomparison Project (CMIP6). Here we describe the main characteristics of this novel earth system-based prediction platform, including the methodological steps for obtaining initial states to produce forecasts. We evaluate the entire C3PS initialisation procedure with the most up-to-date observations and reanalysis over 1960-2021, and we discuss the overall performance of the system in the light of the lessons learnt from previous and actual prediction platforms. Regarding the forecast skill, C3PS exhibits comparable seasonal predictive skill to other systems. At the decadal scale, C3PS shows significant predictive skill in surface temperature during the first two years after initialisation in several regions of the world. C3PS also exhibits potential predictive skill in net primary production and carbon fluxes several years in advance. This expands the possibility of applications of forecasting systems, such as the possibility of performing multi-annual predictions of marine ecosystems and carbon cycle.

Aquatic vegetation alters the hydrodynamics of natural waters, such as rivers, lakes, and estuaries. Plants can generate turbulence that propagates throughout the entire water column, which affects gas transfer mechanisms at both air-water and water-sediment interfaces, driving changes of dissolved oxygen (DO), an important indicator of water quality. We conducted a series of laboratory experiments with rigid cylinder arrays to mimic vegetation using a staggered configuration in a recirculating race-track flume. Walnut shells were chosen as the sediment substrate, which interacts with DO in water. 2D planar Particle Image Velocimetry was used to characterize the flow field under various submergence ratios, highlighting the effect of vegetation on turbulence quantities. Gas transfer rates were determined by measuring the DO concentration during the re-aeration process based on the methodology proposed by the American Society of Civil Engineers. Our data provide new insight on Air-Water-Vegetation-Sediment interactions in streams as a function of submergence ratio, array density, and flow turbulence. A modified surface renewal model using turbulence production as an indicator of gas transfer efficiency is used to predict surface gas transfer rates. A delayed time of re-aeration between the bulk and the near-bed region was observed and varies with flow velocities and submergence ratios, which controls the oxygen flux from water to sediment. Future studies are required to investigate the cause of the delayed time to incorporate sediment oxygen demand in a substrate-to-surface transfer model.

Around the world there is a consensus of the importance of knowladge the ice thickness distribution and total ice volume for estimating future glacier change for both glaciological and hydrological application, especially the Polar Regions for being important regulators of the global climate system and response rapidly to climate change. This work aimed to estimate the ice thickness distribution and total ice volume of Znosko glacier (King George Island, Antarctic Peninsula) using Glacier Bed Topography (Glabtop) model and ground data. Ground penetrating radar (GPR) measurements were collected in two field campaigns. GPR measurements were used to validate, calibrate (shape factor) and improve the accuracy of the ice thickness distribution model. The ice thickness (GPR) distribution is not uniform over the glacier, which varies from 0 to 155 m. The total stored ice volumen of Znosko glacier in 2020 is estimated to be 0.11 km3. Additionally, the model bounded areas in the bedrock topography below sea level, as a sign of future lake formation. Sensitivity analysis reveal that Glabtop model with a shape factor (f) = 0.5 gave best results when compared with ground penetrating radar (GPR) data measured on Znosko glacier and has an uncertainty range of ± 20.5%.

Urban population in sub Saharan Africa (SSA) is rapidly growing. While only 30% of its population lived in urban centers in 2000, this figure will reach 60% by year 2050. Urban energy demand is closely tied to forest degradation. Charcoal is the main source of cooking fuel for eighty percent of African urban households and its overall consumption is expected to rise by 2040. Charcoal production is already the main driver of forest degradation in SSA. REDD+ guidelines encourage countries to identify and describe individual activities and drivers causing forest degradation as an initial step to define suitable methods for measuring and monitoring and formulate appropriate strategies and policies. Yet, forest degradation associated to charcoal production remains largely under reported. Charcoal production results in partial removals of forest cover that do not necessarily involve significant variations of the spectral signal. As a consequence, efforts to monitor forest degradation associated to charcoal production with medium resolution data has proved elusive. We present initial results of our effort to monitor and quantify carbon emissions from forest degradation due to charcoal production in SSA. Our work combines time series of multi sensor medium (20 – 30m), high (2m) and very high (0.5m) spatial resolution sensors with field data to characterize the spatial and temporal dynamics of charcoal production in charcoal production sites across SSA. The integration of these datasets provides the means to map, monitor and measure charcoal kilns, and subsequently quantify the magnitude and intensity of aboveground biomass removals associated to charcoal production at a level of detail and precision not reported previously. Our initial results reveal that charcoal production accounts for a larger share of greenhouse gas emission than previously reported, highlight its negative impacts on the ecosystem, and question the long-term sustainability of charcoal production under current and future urban energy demands. This work is a first step towards the development of a monitoring, reporting and verification system specific to forest degradation in the SSA context.

A document by Lina Garcia-Suarez. Click on the document to view its contents.

It is recapitulated that the gravitational potential energy that is conserved along the neutral surfaces needs two terms, one from buoyancy and the other from gravity. I also show a mathematical identity for the time change of this gravitational potential energy which can be interpreted as exchange of energy amongst kinetic, internal, and gravitational potential forms. Movements along the neutral surface conserve the gravitational potential energy and it is shown that not only conversions into and out of the gravitational potential energy balance, but that each of the conversion terms is zero.

Statistical classifications and machine-learning-based predictive models are increasingly used for environmental data analysis and management. There now exist numerous classifications on the same topic but applied to different regions or spatial scales, such as geomorphic classifications. However, no quantitative meta-analysis framework exists to compare and reconcile across multiple classifications. To fill this gap, we jointly characterize statistical classifications and predictions by combining information theory and machine learning in three novel ways by: (i) measuring the degree of discriminatory information underlying a statistical classification; (ii) estimating the stability of the learning process with tuning entropy; and (iii) leveraging the sequential coarse-graining of information inherent to deep neural networks but absent from traditional machine learning models. This framework is applied through a benchmark of 59 millions models on a unique example of a single statistical classification methodology applied to nine different regions of California, USA. Regional results show that random forest consistently outperforms deep neural networks. In addition, a correlation analysis between regional characteristics, the level of discriminatory information of each classification, and the performance in statistical learning explains variations in performance and reveals the decisive role of the spatial scale of classification outputs. Because such a spatial scale is itself linked to the common situation of limited field sampling, directly comparing findings from statistical classifications and associated predictions appears seldom justified. A more desirable avenue to compare findings lies in combining data underlying statistical approaches in an interpretable and justifiable environmental data science.

We combine earthquake spectra from multiple studies to investigate whether the increase in stress drop with depth often observed in the crust is real, or an artifact of decreasing attenuation (increasing Q) with depth. In many studies, empirical path and attenuation corrections are assumed to be independent of the earthquake source depth. We test this assumption by investigating whether a realistic increase in Q with depth (as is widely observed) could remove some of the observed apparent increase in stress drop with depth. We combine event spectra, previously obtained using spectral decomposition methods, for over 50,000 earthquakes (M0 to M5) from 12 studies in California, Nevada, Kansas and Oklahoma. We find that the relative high-frequency content of the spectra systematically increases with increasing earthquake depth, at all magnitudes. By analyzing spectral ratios between large and small events as a function of source depth, we explore the relative importance of source and attenuation contributions to this observed depth dependence. Without any correction for depth-dependent attenuation, we find a systematic increase in stress drop, rupture velocity, or both, with depth, as previously observed. When we add an empirical, depth-dependent attenuation correction, the depth dependence of stress drop systematically decreases, often becoming negligible. The largest corrections are observed in regions with the largest seismic velocity increase with depth. We conclude that source parameter analyses, whether in the frequency or time domains, should not assume path terms are independent of source depth, and should more explicitly consider the effects of depth-dependent attenuation.

A submesoscale-permitting global ocean model is used to study the upper ocean turbulence. Submesoscale processes peak in winter and, consequently, geostrophic kinetic energy (KE) spectra tend to be relatively shallow in winter (∼k-2) with steeper spectra in summer (∼k-3). This seasonal transition from steep to shallow power-law in the KE spectra indicates a transition from quasi-geostrophic (QG) turbulence in summer to pronounced surface-QG-like turbulence in winter. It is shown that this transition in KE spectral scaling has two phases. In the first phase (late autumn), KE spectra show a presence of two spectral regimes: ∼k-3 scaling in mesoscales (100-300 km) and ∼k-2 scaling in submesoscales (<50 km), indicating the coexistence of QG, surface-QG, and frontal dynamics. In the second phase (late winter), mixed-layer instabilities convert available potential energy into KE, which cascades upscale leading to flattening of the KE spectra at larger scales, and k-2 power-law develops in mesoscales too.

The Yangtze River Valley (YRV) experienced an unprecedented heatwave in midsummer of 2022, but the detailed physical processes involved in the influence of anomalous large-scale atmospheric circulation on the heatwave remain unknown. Here, we show that the positive meridional gradient of anomalous atmospheric moisture at the middle-lower troposphere and associated extreme dry air advection over the YRV are key prerequisites for the formation of the 2022 YRV heatwave. The 2022 YRV heatwave is dominated by the interannual variability, which contributes 72.7% to the total temperature anomalies. Diagnosis of the surface heat budget equation indicates that the surface cloud radiative forcing is the most important process in driving the 2022 YRV heatwave, which is dominated by the positive surface short-wave cloud radiative forcing associated with the suppressed precipitation and the middle-low clouds. The suppressed precipitation is induced by the vertical dynamical processes of anomalous moisture advection caused by the anomalous descending flows over the YRV, which are driven by the negative advection of anomalous latent heat energy by climatological meridional wind (anomalous dry air advection) according to the atmospheric moist static energy equation. Simulations from the Lagrangian model FLEXPART further indicate that the moisture anomaly over the north of YRV is mainly originated from the surface evaporation in the YRV, implying that there is a positive land-air feedback during the life cycle of the YRV heatwave. Our study adds a perspective to the existing mechanism analyses of the 2022 YRV heatwave to serve accurate climate prediction and adaptation planning.

Studies of the dissolved and sediment composition of global rivers provide crucial insights into the relationship between climate, weathering, and landscape dynamics. The chemical composition of suspended riverine sediment constrains contemporary erosion and chemical weathering, and is critical to the interpretation of sedimentary records, such as continental shelf deposits. Here we present suspended sediment flux and chemical composition data from the Irrawaddy (Ayeyarwady) and Salween (Thanlwin) rivers in Myanmar, from samples collected in the wet and dry seasons in 2017-2018. Analyses of major element, 87/86Sr, and εNd composition are combined with a multi-step leaching protocol to determine the ccompositions of silicate, carbonate, and iron oxide components in the bulk sediment. Sources of the organic matter are determined by analyses of carbon and nitrogen isotopes. We show that, as in other large rivers, suspended sediment concentration and composition in the river channel is depth-dependent due to hydrodynamic sorting. Depth profile sampling is therefore required for the complete characterization of the sediment flux and composition in these rivers. Having accounted for hydrodynamic sorting, we demonstrate that there are key differences in the composition of the Irrawaddy and the Salween rivers at the mouth. We further show how the composition of the Irrawaddy sediments evolves along a downstream transect from the northern headwaters to the delta, due to inputs from tributaries draining distinct lithologies and the continued weathering of sediments transported through the floodplain. Finally, we discuss the implications of our findings for the interpretation of offshore sedimentary deposits and for quantifying the regional chemical weathering and carbon budgets.

Characterizing spatial heterogeneity is fundamental in numerous areas, yet defining spatial patterns often depends on qualitative assessments or a priori knowledge. Lacunarity analysis is a popular occupancy-based method for identifying relevant length scales in spatially heterogeneous systems. From lacunarity, we identify the existence of a point which encapsulates the spatial heterogeneity of a given system. This value satisfies the conditions for the lacunarity cutoff function and forms the basis of a heterogeneity index. We evaluate the behavior of both parameters in monofractal, clustered, and periodic systems. In addition, we demonstrate the broad utility of our approach to the scientific community by classifying the spatial heterogeneity of fractured sea ice and comparing our findings to existing measures. The heterogeneity index produced a linear correlation with the area fraction of open ocean to ice with an R2 of 0.967.

Pyrrhotites are a class of geologically important nonstoichiometric iron sulfides with the general composition of Fe1-xS (0 ≤ x ≤ 0.125) and are found in a variety of intergrown polytypes, conventionally separated into the antiferromagnetic hexagonal and the ferrimagnetic monoclinic (4C) varieties. Both structures undergo magnetic phase transitions, where antiferromagnetic polytypes display the λ-transition at ~490K and the 4C polytype shows the Besnus transition at ~30K. However, recent studies have shown the relationship between pyrrhotite polytypes and their magnetic behavior to be more complex and new non-monoclinic polytypes (e.g., 3C) have been described that are also capable of retaining a spontaneous magnetization at room temperature. These advances raise the level of detail needed for the characterization of pyrrhotite in rock magnetic and paleomagnetic studies. This study demonstrates the utility of combining X-ray diffraction data collected as a function of temperature with low- and high-temperature magnetic measurements to characterize natural samples. We analyze two natural samples that contain mixtures of 4C, 5C, and 6C polytypes and describe how their polytypes and magnetic properties vary as a function of temperature across the λ-transition and how and when pyrite and greigite form. We also report the effect on natural samples of an annealing protocol commonly used to elevate the concentration of the 4C polytype in synthetic samples and found that annealing instead transformed some antiferromagnetic pyrrhotite into a form whose diffraction pattern most closely resembles the 4C polytype and displays room temperature spontaneous magnetization, but lacks the characteristic Besnus transition.