In the last decades, volcano monitoring capabilities have increased enormously, thanks to geochemical and geophysical airborne and surface measurements that have steadily improved their accuracy and time resolution. Such a wealth of data is routinely used to track volcanic unrest and eruption evolution, although precise causative links with underground processes are often missing. Modeling of magmatic and volcanic systems has also leaped forward, thanks to the increased availability of computer power and development of numerical models. Capturing the complexity of magmatic system evolution at all scales is nonetheless still a challenge: the crystal- and bubble-size processes need to be taken into account in order to resolve the reservoir-scale dynamics detected by the monitoring networks. We have developed a robust numerical model to solve the thermo-fluid dynamics of magmatic mixtures, that includes pressure-, temperature and melt composition-dependent, locally (space-time) defined properties and constitutive equations of multi-component magmas. The model has been applied to a variety of scenarios related to magma dynamics in underground volcanic systems, including magma arrival from depth into shallow reservoirs. Model results include the space-time evolution of density, pressure, velocity and composition within the domain, that can be used as sources of synthetic geodetic and seismic datasets, akin to those recorded by monitoring networks. The synthetic and monitoring time series can thus be compared and their similarities can be exploited to detect the underground dynamics causing well-defined time and spectral patterns: we are building a physically sound reference for the interpretation of volcanic unrest signals and their relationships with the deep magma dynamics. This approach has been successfully used to detect shallow magma arrival from strainmeter records at Campi Flegrei during the ongoing unrest phase; it is also being applied to evaluate how precise gravity surveys at Mount Etna can help in detecting any shift in the eruptive sequence.

Phosphorus (P) is hypothesised to be the main nutrient limiting forest productivity in tropical forests, but more recent evidence suggests that multiple nutrients could regulate forest functioning. Root functional trait expression represents a trade-off between maximising the acquisition of limiting resources and minimising root tissue construction and maintenance. Therefore, if the limiting soil nutrient supply is increased, plant investment in root biomass and nutrient uptake strategies should decrease. To test this hypothesis we investigated how fine root traits associated with nutrient acquisition responded to large-scale nutrient additions of nitrogen, phosphorus and cations in a slow-growing mature tropical forest established on low fertility soils in the Central Amazon. To evaluate short-term responses to nutrient addition 6 months after fertilisation commenced, we sampled young fine roots (<2mm diameter), measuring root biomass and productivity, root morphological traits (root diameter, specific root length, specific root area and root tissue density) and root phosphatase enzyme activity. We hypothesised that if tropical forests are P limited, responses to P addition would be strongest, resulting in i) a decrease in root production; ii) a shift in root morphology from acquisitive to more conservative traits by increased root diameter and decreased specific length and area and iii) decrease in the investment in phosphatase enzyme. As expected, root phosphatase activity decreased by ~13% with P addition. Among the root morphological traits, root diameter increased, mainly for the 0-10 cm soil layer, with the addition of cations and P, but there were no significant effects on other root morphological traits. Contrary to expectations, root productivity was >50% higher in plots where cations were added, with no effects of P addition. Although we found support for the hypothesis that P limits some aspects of plant functioning in this Central Amazon forest, the results also suggest that cations could play an important role in controlling the expression of root traits. We conclude that multiple nutrients may limit belowground process in Central Amazon forests and that even slow-growing tropical forest can respond very rapidly to changes in soil nutrient availability.

A series of survey protocol for deep sea polymetallic sulfides using not electromagnetic waves but acoustic, electromagnetic, gravity and magnetic, and geochemical methods has been proposedby Japanese national project, the Next-Generation Technology for Ocean Resources Exploration (Zipangu in the Ocean Program) launched in 2014, which enables effective and efficient explorations. The first version of the survey protocol has been made to determine the potential polymetallic sulfides areas using detailed scientific researches on their genesis. Deep sea polymetallic sulfides are produced beneath the sea floor because of the interaction between sea water and sub-bottom hydrothermal solutions driven from volcanic magma, indicating that they must be discovered at mid-ocean ridges, back-arc and arc regions, and hot spots. Recent studies further show that back-arc basins such as Okinawa Trough probably produce deep sea polymetallic sulfides bodies much larger than other regions (Tsuji et al., 2012). Some interesting topics on the survey protocol which further constrains the candidate areas, include acoustic, natural electric potential and seismic survey methods, that are 1) the shipboard Multi Beam Echo Sounding (MBES) survey that can acoustically detect hydrothermal plume including abundant CO2 for regional survey (100km2 <), 2) continuous self-potential measurements with tremendously low noise in the sea water that can identify individual sulfide deposit for semi-detailed survey (10km2- 100km2), and 3) state of the art vertical cable seismic (VCS) system which enables to delineate detailed structure of hydrothermal ore deposit body for detailed survey (< 10km2). These three methods, in combination with other geophysical and geochemical surveys, have been turned out to be effective and efficient for surveying deep sea polymetallic sulfides. Multiple types of data would be integrated into a single visualization to gain a clear view of the structure of the sulfides. Integrating core lithologies obtained via drilling with the seismic reflection structure data has been found to show the seismic velocity structure along with the extent of the polymetallic sulfides bodies.

Drilling by the Oman Drilling Project provided a unique opportunity to access partially-serpentinized harzburgite and dunite. These are in contact with alkaline fluids in a subsurface environment that support a microbial ecosystem. In concert with studies of the rock-hosted microbial community, we are characterizing the mineralogy and petrology of the serpentinized mantle rocks that host this ecosystem. Samples of whole-round core were collected and preserved every 10 m from 3 boreholes and split into paired subsamples for microbiology and mineral characterization. Thin sections were analyzed with a petrographic microscope to complete mineral abundance estimations and interpret textural relationships. Raman spectroscopy was conducted on the thin sections to reveal structural/compositional data about mineral phases. Powders were prepared for XRD analysis for quantitative phase identification. The main rock types are altered harzburgite and dunite, and altered veins of gabbro or pyroxenite occur at certain depths. All of the cores have experienced multiple episodes of serpentinization. The observed mineral assemblages include relict olivine, pyroxene and abundant secondary serpentine, brucite, iron sulfide and andradite-grossular garnet. The assemblages are generally expected from partial serpentinization of peridotite, but the widespread distribution of garnet was particularly surprising. Over 50% of the samples contained sufficient garnet to be detected by XRD. Optical and Raman analyses show that garnet occurs in many textural contexts, notably inside mm-scale, late-stage serpentine veins. Andradite garnet in serpentine veins similar to those found here are likely to have formed during serpentinization at temperatures below ~200°C [1,2]. Incorporating Fe3+ into the andradite component could facilitate H2 production, a potent energy source for microbial metabolisms [2]. Its high abundance may provide key insights into H2 production and habitability during late-stage serpentinization of the Oman ophiolite. [1] Ménez et al. (2018) LITHOS DOI: 10.1016/j.lithos.2018.07.022 [2] Plümper et al. (2014) Geochimica et Cosmochimica Acta 141 (454-471).

The Foundation Spatial Data Framework (FSDF) is a framework of ten national authoritative geographic data themes that supports evidence-based social-economic decision making across multiple levels of Australian and New Zealand government agencies, industry, research and the community. The AAA data management principles (Authoritative, Accurate and Accessible), articulated for FSDF, are easily translatable to the FAIR Principles and applied to ensure: • Ability to Find data through rich and consistently implemented metadata; • Access to metadata and data by humans and machines while practicing federated data management within trusted data repositories; • Interoperability of metadata and data through adoption of common standards and application of best practices; and • Reusability of data by capturing licencing constraints and information about its quality and provenance. The Location Information Knowledge Platform (LINK) was developed in 2016 as a digital catalogue of FSDF content. This governed, online, dynamic, analysis and discovery tool was designed to enhance the discovery of FSDF datasets, support work planning and indicate the legal frameworks, agency priorities and use case associated with FSDF data. More than 73 Australian government agencies and commercial organisations use this Platform. Current work includes: • Building common high-level and individual lower-level information models (ontologies) for the FSDF and each dataset; • Development of a new architecture for persistent identifiers and identifier incorporation in the datasets; • The ISO 19115-1-based Australian and New Zealand Metadata profile and best practices user guides; and • Testing new workflows for metadata and data governance and integration utilising a set of common cloud-based infrastructure. On realisation, the FSDF will become a necessary component of spatial socio-economic decision making across Australian and New Zealand government agencies and the private sector. FSDF will encourage cross-sector partnerships and enable seamless access to authoritative spatial data across organisational and jurisdictional boundaries, thus contributing to economic growth, improved public safety, meeting legal and policy obligations and sustaining business needs.

JavaScript Object Notation (JSON) is an increasingly popular text format for data exchange. netCDF encapsulates the Common Data Model (CDM) and a binary format for machine-independent and network-transparent storage of scientific data and metadata. Previous netCDF-to-JSON translators have been custom solutions with incomplete features, or based on more complex formats than the CDM. Here we describe a flexible JSON format that describes any classic or extended format netCDF dataset. This format, called NCO-JSON, expresses the richness of the CDM and increases interoperability between web services and netCDF data. NCO-JSON requires no reserved keywords and so is completely compatible with all netCDF datasets. It allows for selectable levels of fidelity to the original data and metadata. The most concise and human-legible form of NCO-JSON is also lossy. By design it distinguishes only the three atomic JSON datatypes (float, string, and int). This suffices for many purposes yet cannot guarantee bit-for-bit reproducibility of many netCDF datatypes, especially in round-trip translations. NCO-JSON uses a more complex object notation to encode the additional type information required to reproduce netCDF datasets with full fidelity. We present the rules and design of the NCO-JSON format, show results with real-world datasests, quantify the space advantages vs. alternate formats (both JSON and XML), and discuss corner cases and possible extensions.

VISAGE (Visualization for Integrated Satellite, Airborne, and Ground-based data Exploration) aims to provide visualization and analytic capabilities for diverse datasets in an interactive user interface. Proof-of-concept use cases are centered around the Global Precipitation Measurement (GPM) mission’s Ground Validation (GV) program, which provides a wealth of intensive, coincident observations of atmospheric phenomena from a wide variety of ground-based, airborne and satellite instruments. These data have diverse temporal and spatial scales, variables, and data formats and organization. Key technical challenges include: 3D data rendering and visualization of multiple diverse datasets on a web-based platform 3D data interrogation via map user interface Temporal alignment of data with diverse time scales and resolutions Computations on data fields across instruments and platforms To address these issues, the VISAGE project is working with cloud-native serverless technologies to render data as 3D Tile Point Clouds for display in the Cesium geospatial 3D global mapping platform.

Climate models at high resolution (~25 km horizontal grid spacing) can permit realistic simulations of tropical cyclones (TCs), thus promising the investigation of these high-impact extreme events under present and future climates. On the global scale, simulations with the Community Atmosphere Model version 5 (CAM5) present a reasonable TC climatology under prescribed present-day (1980-2005) sea surface temperature (SST) and greenhouse gas (GHG) forcing. However, for the disaster-prone western North Pacific (WNP) region, biases in TC genesis frequency and location persist across various configurations. The biases under-represent the basin’s share in global TC climatology, complicating the fidelity of future projections. This study addresses these model biases in WNP by evaluating the large-scale environmental controls of TC genesis in CAM5 with two aerosol configurations. Across the two configurations, the lack of mid-level moisture is consistently identified as the leading cause of the deficit in simulated WNP TC genesis. This lack of mid-level moisture in WNP TC main develop region is potentially linked to previously identified deficits in Pacific warm pool precipitation at high horizontal resolution in CAM5, as well as biases in the East Asian Summer Monsoon circulation and moisture transport. Additional CAM5 simulation experiments will explore the effect of moisture nudging on the large-scale environment and subsequent TC genesis, tracks, and intensity development. For a chosen year, simulations covering WNP peak TC season (July - October) under otherwise identical forcing (SST, GHG etc.) will be run with and without nudging the specific humidity field towards MERRA-2 reanalysis. The insight into the biases of basin-scale TC simulation under the present climate and potential improvements will help reduce the uncertainty in future-climate projections, in the interest of disaster risk management.

Newer satellite platforms, such as NISAR, are poised to produce huge amounts of data that require large computational resources. Currently, researchers typically download datasets for analysis on local computer resources. This paradigm is no longer practical given the volumes of data from new sensing platforms. While cloud computing services offer a potential solution for accessing and managing large computational resources, there remains a significant barrier to entry. Levering cloud services requires users to: navigate new terminology without appropriate documentation; optimize settings for services to reduce costs; and maintain software dependencies, upgrades, and allocated hardware resources. A more accessible approach for migrating earth scientists to the cloud is needed. To address this problem, we are developing the open source Python library PODPAC (Pipeline for Observational Data Processing Analysis and Collaboration), with the goal of helping to address NASA’s rapidly growing observational data volume and variety needs. PODPAC enables earth scientists to seamlessly transition between processing on a local workstation (their current paradigm) to distributed remote processing on the cloud. It does this by leveraging a text-based JSON format automatically generated for any plug-and-play algorithm developed using PODPAC (e.g., in a Jupyter Notebook). This text format describes data provenance, and is used in RESTful web requests to preconfigured PODPAC cloud deployments, allowing scalable, massively distributed processing. We demonstrate the seamless transition to the cloud by developing a simplified soil moisture downscaling algorithm in Python using PODPAC. Data for this algorithm uses NASA Soil Moisture Active Passive (SMAP) sensor retrieved from the National Snow and Ice Data Center using OpenDAP, and fine-scale topographic data retrieved via Open Geospatial Consortium (OGC) Web Coverage Service (WCS) calls. We then use a serverless AWS Lambda function to run the same algorithm using the automatically-generated text format. Our generic preconfigured environment can handle a wide variety of processing pipelines, and scale up to 1024 parallel processes. This approach enables incremental adoption of cloud services by researchers, significantly lowering the barrier to entry.

The Aburrá Valley is a narrow highly complex mountainous terrain located in the Colombian Andes. Due to topographical features of the region, and the tropical setting, the meteorological variability is very high and in specific periods of the year limit the atmospheric pollutant vertical dispersion, resulting in high concentrations within the valley. The presence of prevalent low-level clouds in these periods reduce incoming solar radiation to the surface thus diminishing surface sensible heat flux to the lower atmosphere. Therefore, the spatial distribution and temporal variability of cloud coverage play a crucial role in the surface energy balance in the region. Cloud variability is also relevant to study the local hydrological cycle and long-term climate variability and change. The most widespread techniques for cloud observations are human observations, which strongly depends on the objectivity and commitment of the observer, and satellite observations, that has the disadvantage of having a relatively low temporal and spatial resolution for some applications. This research focuses on the implementation of an operational system for in-situ clouds detection based on a ground-based network of whole-sky visible cameras. The operational system uses computer vision techniques and image classification algorithms such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN) among others to obtain. The algorithm starts with a group of training images with set attribute tags such as cloudy and cloud-free skies, to identify and calculate the percentage of clouds coverage in untagged images. The methodology then projects the images from the natural polar coordinates system of whole-sky cameras to 2D cartesian coordinates. Following the reprojection, overlapping images from each camera are combined using a panorama stitching technique to generate a single regional cloud fraction map. Clear-sky combined direct and indirect incoming solar radiation are adjusted using the regional cloud map to generate an estimate of the spatial distribution of cloud-forced incoming solar radiation. Cloud height from ceilometers and in-situ pyranometer measurements of incoming solar radiation provide the additional required information to generate the radiation maps.

The influence of clouds on photochemistry remains a significant uncertainty in global chemistry models. Variability in cloud fraction, morphology, phase and optical properties provides significant challenges to models with horizontal resolutions that far exceed the scale of most clouds. Measured photolysis frequencies derived from the Charged-coupled device Actinic Flux Spectroradiometers (CAFS) on board the NASA DC-8 during the Atmospheric Tomography (ATom) mission in 2016 provide an extensive set of statistics on how clouds alter the photolytic rates throughout remote ocean basins. Here we focus on north and tropical pacific transects during the first deployment (ATom-1) in August 2016 including regular profiles through cloudy, partly cloudy and clear conditions. Nine global chemistry–climate or –transport models provide similar statistics on J-values for regional domains encompassing the measured flight path. The statistical picture of the impact of clouds on J-values emerges through the distribution of the ratio of the cloud influenced models and measurement to corresponding cloud free model runs (J-cloudy/J-clear). The models all reproduce general patterns of enhancement above and shading below cloud, but diverge in distribution patterns and clear sky prevalence.

There is growing concern that the quality of science education today is not adequate to provide students with the level of scientific literacy they will need to be effective citizens in the 21st century. To address this concern, Educational Seismology Projects (ESPs), such as the Texas Educational Seismic Project (TX-ESP) and the Boston College (BC) Educational Seismology Project (BC-ESP) promote inquiry-based learning and investigative methods in Earth science classrooms, at public events, and in after-school activities. TX-ESP is a 501c3 Non-Profit Public Charity which focuses on educational seismology outreach and collaborative research with BC-ESP. BC-ESP is operated by BC’s Weston Observatory, a research and science education center of its Department of Earth and Environmental Sciences. Our common goal is to enhance science education by offering opportunities for students, teachers, and the public to be directly involved with research scientists. Through our educational partnerships, we make a difference in the lives of citizens and students of all ages and in differing socio-economic circumstances by engaging in hands-on learning opportunities and teaching science as it is actually practiced. The science of seismology forms an excellent foundation for this endeavor because: (1) it is an interdisciplinary science that requires integration of many STEM concepts, and (2) it teaches how the natural environment impacts our everyday lives. A particular advantage of this collaboration is the locations of our ESPs. BC-ESP is located far from plate boundaries, where students don’t tend to record as many earthquakes as they do at TX-ESP, which is much closer to plate boundaries and thus records more earthquakes, with stronger signals, than BC-ESP. When BC-ESP students see the TX-ESP recordings (on their own web-interface), they often find more subtle signals of the same earthquakes on their New England seismograms. This provides teachable moments for students at both locations to investigate seismic wave propagation and attenuation, as well as differences in seismicity between intraplate and plate boundary environments. Also, TX-ESP students leverage a high-quality seismology curriculum through BC-ESP’s research experience and its successful, long-term ESP presence in New England.

Studies of interior air exposures to various human and non-human components has largely been restricted to industrial exposures for the purpose of regulation. In contrast, little attention has been paid to exposure at the residential scale, where people spend much of their day and may be exposed to particulate sources ranging from known toxins, such as lead, arsenic, and asbestos, to human-produced chemicals of yet unknown toxicity, such as flame retardants. To capitalize on experience with citizen science initiatives as they pertain to environmental health, researchers formed an international network called 360 Dust Analysis, which provides guidance on citizen science and interior dust collection, as well as research tools to examine dust through analysis in regional labs. We present initial results from the July 2018 launch of this program in the USA, called DustSafe USA and operated under approved human subjects protocols by Indiana University (http://www.urbanhealth.iupui.edu/). We launched via multiple media strategies, including an extended television news segment, an article in several Indiana newspapers, appearances in several statewide radio shows, and via a widely distributed press release. As of this abstract submission, well over 300 queries were received, and after only two weeks of the launch the lab has received nearly 100 dust samples. Participants are largely from central Indiana where most of the media play occurred, but samples have also come from all over the country. We will present geochemical and compositional results from the dust analysis, but perhaps more importantly we will discuss how citizens were engaged, how the funding model for such efforts might be developed, and the general approach to research translation and citizen science.

A growing number of initiatives in disparate scientific domains arose in the last decade, with the common goal of clustering data and services resources across Europe and make them integrated, open, sharable and available through infrastructures built according to FAIR principles[1]. In this context, the European Plate Observing System (EPOS[2]), now in its Implementation Phase and soon getting to the status of ERIC, is a long-term plan to facilitate integrated use of data, data products, software, services (DDSS) from distributed research infrastructures in Europe in the Solid Earth Domain. Its innovation potential consists in the opportunity of integrating distributed heterogenous resources from European, National and institutional resource providers, and making them available in one single environment. EPOS technical architecture has three main layers: a) National Layer of data providers, that provide access to DDSS; b) community-specific, European-Wide Thematic Core Services (TCS) layer, that collect and integrate DDSS from specific sub-domains and make them available at European level; c) EPOS Integrated Core Services (ICS) system where the integration of DDSS occurs. The architecture relies on three main concepts: Metadata: it is fundamental for describing assets and resources managed by ICS. A twofold approach was used for metadata: at metadata management level, the CERIF model was used for storing all information within the system; at metadata transfer level, an extension of DCAT-AP[3] was created (EPOS-DCAT-AP[4]) to facilitate TCS metadata collection. An architectural approach based on microservices that ensures scalability, flexibility and system interoperability. A harmonization process, that focused on technical aspects like data formats and protocols to access DDSS, but also required intra-domain work on semantic interoperability that includes adoption of common standards and vocabularies. We will discuss these topics and show a demonstration of the ICS prototype. [1] https://www.force11.org/group/fairgroup/fairprinciples [2] https://www.epos-ip.org/ [3] https://joinup.ec.europa.eu/solution/dcat-application-profile-data-portals-europe [4] https://github.com/epos-eu/EPOS-DCAT-AP/

The Space Physics Data Facility (SPDF) has developed and/or leveraged standardized self-describing data formats, metadata for datasets and parameters, time conventions, and dataset and filenaming conventions that enable effective data analysis and browsing using generic easy-to-use software and web services. Software and services include SPDF’s CDAWeb , and external tools such as Autoplot and SPEDAS IDL library. Standards and conventions include: datasets and filenaming and , the CDF scientific data format (including its new Python library ), the ISTP/IACG/SPDF Guidelines for global and variable attributes , time variable types , and the SKTeditor metadata creation tool . The SPASE standards for describing datasets for easy searching are crucial to the Heliophysics Data Portal .

In the paper summarized by this presentation we trace how maps aided the conceptualization and ensuing debate over one of the first inherently spatial and multiscale environmental issues, acid rain. Through interviews with early acid rain ecologists and a critical cartographic analysis of printed maps, we show that mapping was central for constructing this environmental issue. Rather than static representations of a scientific reality, acid rain maps were interconnected and relational processes that operated within political and economic discourses. Accordingly, we suggest that increased critical engagement and participation with the process of mapping can productively aid dialog on many issues, including other current multiscale environmental issues, such as climate change.

Over recent years, remote sensing of sea ice has advanced at a rapid pace. However, there are inherent limitations in the ability of existing space and airborne sensors to observe changes in the properties of near-shore sea ice, especially over short (hourly) time scales. This information is of critical importance to the livelihood of local communities and to meteorologists who depend on knowledge of near-shore ice conditions for weather prediction. The use of near-real-time data from coastal seismic arrays promises to advance coastal ice observations by measuring the amplitude of background seismic noise, known as microseism. The microseism signal is generated by interactions between oceanic waves, the ocean floor, and the shoreline. Previous studies have shown that along polar coastlines the microseism is modulated by the presence of sea ice. In this feasibility study, we explore the use of power spectral density (PSD) measurements from the Utqiagvik station of the EarthScope Transportable Array (TA) to provide information about sea ice conditions off the northern coast of Alaska. PSD signals are compared with daily estimates of near-shore ice extent and concentration within the Beaufort and Chukchi seas. These are derived from satellite passive microwave radiometer data as well as visible and short-wave infrared imagery from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) instruments. The amplitude of microseism at a frequency near 1 Hz is statistically correlated with ice coverage to determine if microseismic signals from a coastal station can be used to reliably identify particular ice events, including the onset date of summer melt, fast-ice breakup and formation, and the development of near-shore flaw-leads and polynas. Data from the Utqiagvik TA station is compared with observations from other northern coastal stations to determine if sea ice related microseismic signals are consistent across a range of geological and topographical environments. The expansion of the EarthScope TA seismic network to the Arctic coastline since 2011 presents a developing approach to sea ice observation. In the future it may complement established remote sensing techniques to provide a more complete picture of coastal ice conditions as they evolve.

There is no consensus on the physical mechanisms controlling the scale at which convective activity organizes near the Equator, where the Coriolis parameter is small. High resolution cloud-permitting simulations of non-rotating convection show the emergence of a dominant length scale, which has been referred to as convective self-aggregation. Furthermore, simulations in an elongated domain of size 12228km x 192km with a 3km horizontal resolution equilibrate to a wave-like pattern in the elongated direction, where the cluster size becomes independent of the domain size. These recent findings suggest that the size of convective aggregation may be regulated by physical mechanisms, rather than artifacts of the model configuration, and thus within the reach of physical understanding. We introduce a diagnostic framework relating the evolution of the length scale of convective aggregation to the net radiative heating, the surface enthalpy flux, and horizontal energy transport. We evaluate these length scale tendencies of convective aggregation in twenty high-resolution cloud-permitting simulations of radiative-convective equilibrium. While both radiative fluxes contribute to convective aggregation, the net longwave radiative flux operates at large scales (1000-5000 km) and stretches the size of moist and dry regions, while the net shortwave flux operates at smaller scales (500-2000 km) and shrinks it. The surface flux length scale tendency is dominated by convective gustiness, which acts to aggregate convective activity at smaller scales (500-3000 km). We further investigate the scale-by-scale radiative tendencies in a suite of nine mechanism denial experiments, in which different aspects of cloud radiation are homogenized or removed across the horizontal domain, and find that liquid and ice cloud radiation can individually aggregate convection. However, only ice cloud radiation can drive the convective cluster to scales exceeding 5000 km, because of the high optical thickness of ice, and the increase in coherence between water vapor and deep convection with horizontal scale. The framework presented here focuses on the length scale tendencies rather than a static aggregated state, which is a step towards diagnosing clustering feedbacks in the real world. Overall, our work underscores the need to observe and simulate surface fluxes, radiative and advective fluxes across the 1km-1000km range of scales to better understand the characteristics of turbulent moist convection.