Environmental Engineering Impact Studies

I am thrilled to share our new global dataset on the drivers of forest loss at 1 km resolution, which has been 2+ years in the making! We developed the data using a customized ResNet model trained on a set of samples we collected through visual interpretation of very high-resolution satellite imagery. The model used satellite imagery (Landsat & Sentinel-2) and ancillary data to classify seven driver categories: permanent agriculture, hard commodities (e.g. mining and energy infrastructure), shifting cultivation, logging, wildfires, settlements and infrastructure, and other natural disturbances. This data provides important insights on where tree cover loss is likely to be associated with long-term land use change versus temporary disturbances that may followed by forest regrowth, and can enable targeted solutions to protect and sustainably manage the world's forests. 👉 Read more from our paper, published today in Environmental Research Letters: https://lnkd.in/gxV34-3W 👉 Read a summary of the findings (updated to 2024) here: https://lnkd.in/gghcyVzx 👉 Read our technical blog on GFW here: https://lnkd.in/gBv5ErKU The data is available on: 🌎 Google Earth Engine: https://lnkd.in/gt4t9zhp 🌏 World Resources Institute's Data Explorer: https://lnkd.in/gbVMUzpx 🌍 Global Forest Watch: https://gfw.global/2LUOmIx 🌍 Zenodo (including training + val data): https://lnkd.in/gsn-J9gg I am super proud of this effort and of our team! Radost Stanimirova, PhD, Anton Raichuk, Maxim Neumann, Jessica Richter, Forrest Follett, James MacCarthy, Kristine Lister, Christopher Randle, Lindsey Sloat, Elena Esipova, Jaelah Jupiter, Charlotte Y. Stanton, PhD, Dan Morris, Christy Melhart Slay, Drew Purves, Nancy Harris A great collaborative effort between Global Forest Watch, Land & Carbon Lab, and Google DeepMind, with early contributions from The Sustainability Consortium

Critical minerals are no longer simply natural resources. They are becoming strategic infrastructure. As industries accelerate investment in AI, advanced manufacturing, electrification, semiconductors, and next-generation technologies, access to critical minerals is emerging as a defining component of long-term competitiveness. Mineral reserves do not automatically translate into economic advantage. Extraction capacity, processing capability, infrastructure, investment, governance, and resilient supply chains all influence how those resources create value. For business leaders, this extends well beyond the mining sector.  Many organizations now operate in industries that depend on supply chains built around materials they neither produce nor directly control. Understanding where critical resources originate—and how those ecosystems evolve—is an essential element of long-term strategy and operational resilience. Competitive advantage is increasingly shaped not only by innovation, but by the ability to secure the capabilities and resources that make innovation possible.

Denmark has announced it will plant 1 billion trees and convert 10% of its farmland into forests and natural habitats over the next two decades. With a budget of 43 billion kroner / $6.1 billion, the country aims to reduce fertiliser usage, restore low-lying, climate-vulnerable soils, and expand forested areas by 250,000 hectares. This represents the most significant transformation of the Danish landscape in over a century, with numerous economic and environmental benefits. What are the economic benefits? 1. Job Creation: Large-scale reforestation and land restoration projects will generate employment opportunities in sectors like forestry, environmental management, and sustainable agriculture. 2. Sustainable Agriculture: Reducing fertilizer usage promotes environmentally friendly farming practices, which can lower long-term costs for farmers and mitigate environmental degradation. 3. Climate Resilience: Expanded forested areas act as carbon sinks, reducing climate change impacts. Restoring ecosystems can stabilize agricultural yields and decrease the economic toll of climate-related disasters. 4. Biodiversity and Ecosystem Services: Restored habitats improve biodiversity, which enhances essential ecosystem services such as pollination and water purification, benefiting various economic sectors. 5. Tourism and Recreation: New natural landscapes can boost eco-tourism and recreational activities, contributing to local and national economies. What is the impact of reducing farmland on the economy? Denmark’s decision to reduce farmland is a calculated step toward sustainability, offering both immediate and long-term advantages: • Improved Land Use Efficiency: By targeting marginal or low-yield agricultural lands that require excessive inputs, Denmark reduces resource waste and prioritizes areas with higher ecological value. Farmers may adopt innovative technologies like precision agriculture to maximise yields on remaining farmland. • Economic Diversification for Farmers: Financial compensation helps farmers transition into alternative ventures such as eco-tourism, sustainable timber production, or specialty crop farming. This provides more stable and diverse income streams. • Reducing Soil Degradation: Farmland reduction helps restore soil health and fertility, ensuring long-term agricultural productivity while reducing costs associated with soil erosion and nutrient loss. • Climate Change Mitigation: Reforested areas will sequester carbon, contributing to global climate goals and reducing future economic risks tied to climate impacts. • Balancing Global Food Security: By improving agricultural efficiency and focusing on high-value crops, Denmark can contribute to sustainable global food systems without overproducing low-margin commodities. Learn more: https://lnkd.in/dZx86iUj #economy #reforestation #restoration #land #sustainable #ecosystem

It is time for #NVIDIA, #AMD, #Intel and every other AI hardware vendor to stop hiding the true environmental cost of their products. Billions are being invested in AI hardware, yet we still lack transparent data on embodied emissions, resource use, water intensity and toxicity impacts. NVIDIA (et al) release selective impact assessments designed to meet compliance needs but conveniently exclude everything that matters. A great new study helps to fill some key gaps: “More than Carbon: Cradle-to-Grave Environmental Impacts of GenAI Training on the NVIDIA A100 GPU”. Its got so much valuable info and is worth a read. https://lnkd.in/eSwzd624 Unlike most studies that rely on secondary data, the researchers physically dismantled an NVIDIA A100 GPU ground it up and carried out a full elemental composition analysis. Using that data, they modelled sixteen environmental impact categories across the entire life cycle, covering raw material extraction, manufacturing, model training and end-of-life. The findings are so interesting: 1. Manufacturing is the dominant source of impact a) Manufacturing a accounts for 81.8% of the total climate impact and 80% of fossil resource depletion before it trains a single model. b) 71% of mineral and metal depletion and 94.5% of cancer-related human toxicity impacts occur during manufacturing. c) The copper-heavy heatsink alone is responsible for 91% of cancer related toxicity, 86% of freshwater eutrophication and 91% of land use impacts. d) Semiconductor fabrication at 7nm is a hotspot, with each square cm of silicon requiring significantly more energy, chemicals and water than previous generations. 2. Training is highly energy intensive but not the whole story a) Training GPT-4 on A100s consumed the equivalent of 11,522 people’s annual climate-change budget. b) In Iowa, where GPT-4 was trained, the carbon-intensive grid drives 96.8% of the training climate footprint. c) Focusing on energy efficiency alone will not solve the problem. Operational carbon dominates the impact, but toxicity, water stress and mineral depletion are driven by manufacturing. 3. AI’s material dependency is huge and invisible a) An A100 contains dozens of rare earths & critical minerals including copper, gold, palladium, platinum and tantalum. b) They found a 33% increase in mineral and metal depletion impacts compared with standard LCAs (i.e. secondary data significantly underestimates things). c) Semiconductor fabrication is concentrated in water-stressed regions such as Taiwan, South Korea and Arizona, yet vendors do not disclose water intensity per GPU. This is why hardware vendors must conduct full component level PCF's incl. verifiable embodied impact data. Without transparency we are literally flying blind whilst they make trillions of dollars. This study is an important milestone, but it also shows how little we really know about the environmental impact of AI hardware. Vendors... stop hiding

Can we use #LCA to measure a product system's impact on #biodiversity ❓ The answer is yes❗ - How reliable are these calculations? Well, that is up for discussion. The impact on biodiversity should always be measured in situ by surveying the species richness of and ecosystem and in combination with other techniques usually including local communities' knowledge. - Why do I think so? Because ecosystems are essentially unique everywhere we look, the impact of a substance emission or material extraction from nature (elementary flows) varies from region to region. It is different to perform a given activity in an urban area than in a rainforest. However, in the last decade, new Life Cycle Impact Assessment methods have been developed to account for regional differences in the impact on biodiversity. They typically focus on assessing the impacts of #landuse and land-use change, as these are among the most significant drivers of biodiversity loss. They may quantify impacts in terms of potentially disappeared fractions of species (PDF) over a certain area and time (usually m2/year) or use other metrics to estimate the change in species richness or ecosystem quality. Some of the methods that include approaches to assess biodiversity impacts are: ➖ ReCiPe: a comprehensive LCIA method that includes a model for assessing land use impacts on biodiversity through the PDF metric. It aims to quantify species loss over a certain area and time due to land use. ➖ IMPACT World+Endpoint: This method includes an attempt to integrate biodiversity impacts through several impact categories such as the PDF from freshwater acidification, damage to ecosystem quality from changes in the soil pH, marine acidification, ecotoxicity, land transformation and occupation, water pollution, and water availability. It is one of the most complete. ➖ USEtox: focused on toxicological impacts, includes considerations for ecotoxicity, which indirectly affects biodiversity by assessing the potential toxic impacts on aquatic and terrestrial species. ➖ Land use biodiversity (Chaudhary et al., 2015): recommended by the UNEP-SETAC Life Cycle Initiative: "The indicator represents regional species loss taking into account the effect of land occupation displacing entirely or reducing the species that would otherwise exist on that land, the relative abundance of those species within the ecoregion, and the overall global threat level for the affected species." I love this method because includes regional factors. ➖ Global Biodiversity Score (GBS): not a traditional LCIA method, GBS is a tool developed to help companies assess their impact on biodiversity. Using a common metric, it translates pressures from organizational activities into impacts on biodiversity. We need to think way beyond #carbonfootprint to aim for a #sustainable world. Biodiversity loss is that issue that although highly interlinked with #climatechange, is the actual major environmental issue we face.

WHY NATURE MUST BE INCLUDED IN CORPORATE BALANCE SHEETS For a long time, corporate accounting ignored the obvious: that all economic activity depends on nature. Over the centuries, we have normalized the idea that environmental resources are free, inexhaustible, and therefore irrelevant to financial statements. This omission, however, has become a systemic error. By recording mineral deposits, commodity inventories, or land use rights as assets—while ignoring the ecosystems destroyed to access these resources—we have created an accounting system that conceals risks, distorts profits, and perpetuates an illusion of prosperity. As economist Pavan Sukhdev warned, “we are stealing from the future to fund the present—and calling it GDP.” The deforested forest is not recorded as a loss. The contaminated aquifer doesn’t show up on the balance sheet. The disrupted water cycle is treated as a coincidence. The prevailing logic separates nature and economy, as if the former were a passive backdrop for the latter to unfold. The truth is that more than half of the world’s GDP is moderately or highly dependent on nature’s services, as pointed out in the Dasgupta Review. Fertile soils, clean water, pollination, carbon sequestration, climate stability—these are all essential economic inputs. But by failing to recognize them as assets—and by not recording their degradation as liabilities—companies overestimate their strength and underestimate their risks. Floods, droughts, wildfires, and biodiversity collapses disrupt supply chains, erode reputations, and threaten markets. The World Economic Forum estimates that the loss of nature could cost nearly half a trillion dollars per year. And this cost, for now, remains absent from financial statements. Some companies are starting to move forward. French firm Kering has developed an Environmental Profit & Loss model that assigns monetary value to the environmental impacts of its operations. In Brazil, Natura has for decades incorporated social and environmental accounting methods that recognize the role of the forest and traditional communities as part of its value chain. But these are exceptions in a system still blind to natural capital. Economic flows must operate within the planet’s ecological limits—not outside them. Biomimicry, for example, proposes that businesses learn from living ecosystems. Embracing this logic means abandoning the idea that nature is a stock to be exploited and instead seeing it as a living system with which we must reconcile. More than a technical innovation, including nature in balance sheets is an ethical and strategic step. Ethical, because destroying the foundations of life to generate dividends is a moral contradiction. Strategic, because without nature, there is no future for business. Nature must be included in balance sheets. Corporate accounting can no longer be limited to the language of profit—it must learn to speak the language of life.

The story that synthetic nitrogen represents a technological triumph that saved the world from hunger mischaracterizes its industrial origins. These fertilizers weren't designed for optimal crop nutrition; they emerged from repurposed wartime chemical manufacturing. Haber-Bosch wasn't developed to feed populations—it was engineered to manufacture explosives. After WWII, the chemical industry faced massive production overcapacity and strategically pivoted toward agriculture, creating a new market for existing industrial infrastructure. This transformation wasn't driven by agricultural necessity but industrial pragmatism. Today, 85% of global ammonia production remains tied to fertilizer manufacturing, a direct legacy of post-war industrial strategy rather than a carefully optimized plant nutrition system. The environmental footprint is substantial. Haber-Bosch uses 1-2% of global energy and 3-5% of natural gas, prioritizing industrial efficiency over ecological integration. The entire process is structured around industrial chemistry rather than biological systems. The Green Revolution doubled down on this trajectory, breeding crop varieties to maximize response to synthetic nitrogen inputs and creating a self-reinforcing dependency cycle that marginalized knowledge systems that had sustained diverse agricultural ecosystems for millennia. Ecological consequences extend beyond immediate agricultural impacts: –Accelerated soil carbon depletion that undermines fertility –Decimated soil microbiomes –Nitrogen runoff (1.5M metric tons annually in Mississippi Basin) –Crops with significantly less nutrient density Synthetic nitrogen represents a reductive approach to crop nutrition. These fertilizers deliver nitrogen in forms that disrupt microbial signaling pathways, accelerate nutrient leaching, and create metabolic inefficiencies, compromising nutritional density and plant stress resilience. We now have a greater understanding of the sophisticated biological nutrient acquisition pathways that farmers can harness. Plants have evolved complex microbial relationships that extract and cycle nutrients through ecological rather than industrial processes. The problem isn't technological limitations but institutional inertia and structural bias. Between 1950-2000, a mere 2% of agricultural research funding went to agroecological approaches, systematically marginalizing alternatives while reinforcing industrial paradigms. Research priorities have consistently privileged yield-centric metrics over ecosystem services, soil health, and resilience—creating a narrow technological vision that benefits industrial interests rather than optimizing for long-term agricultural viability. The path forward requires a fundamental reimagination of agricultural systems that prioritizes biological complexity and ecological processes over reductive chemical approaches. This transition represents not just technological substitution but a profound philosophical shift.

Very happy to share that we've now open-sourced our high-resolution (30m) nationwide dataset on Land Use/Land Cover (LULC) and terrestrial carbon storage in Pakistan, covering the period from 1990 to 2020. This dataset accompanies our recent study published in the Environmental Impact Assessment Review journal, providing critical evidence on how urban expansion affects carbon sequestration in Pakistan. ✦ Key Insights from Our Study: ⚬ Exponential Urban Growth: Urban areas in Pakistan have expanded by approximately 1040%, resulting in substantial changes to the country’s landscape. ⚬ Reduction in Carbon Storage: This urbanization has led to a 5% decline in terrestrial carbon storage, posing challenges for climate change mitigation. ⚬ Regional Dynamics: ▸ Emerging Cities on the Rise: Cities like Rawalpindi and Peshawar experienced rapid urban sprawl, primarily converting rangelands (~47%) and agricultural areas (~35%) into urban landscapes. ▸ Afforestation Efforts: While northern afforestation projects have increased forest carbon stocks, there is a marked north-south disparity in carbon storage loss. ▸Land Use Changes: The shift from natural ecosystems to built-up areas highlights the urgent need for sustainable urban planning. ✦ Implications for Pakistan: ▸Climate Change Mitigation: This dataset is essential for understanding carbon storage dynamics, a critical component of strategies to achieve net-zero emissions. ▸Policy Development: It offers valuable insights to support sustainable land-use practices and evidence-based policy-making. ▸Research and Collaboration: Open access enables collaborative efforts among researchers, urban planners, and environmental managers, fostering data-driven environmental management solutions. By openly sharing this dataset, we aim to empower stakeholders to make informed decisions that balance urban development with environmental conservation. Open data promotes transparency, collaboration, and innovation, vital in addressing the complex challenges of climate change and rapid urbanization. 🔗Project Link: https://lnkd.in/eMDQMaHm #pakistan #lulc #landuse #landcover #urban #urbansprawl #urbanization #carbon #carbonstorage #googleearthengine #opensource #remotesensing #geospatial #gis

Dear all, Going live today … in collaboration with the World Resources Institute Land & Carbon Lab and Global Forest Watch, we used AI to map the global drivers of forest loss at 1km resolution - a 100x improvement over previous (10km) state of the art models -- for every year 2001 - 2022. Forest loss is the single largest threat to terrestrial biodiversity; is responsible for c. 15% of anthropogenic CO2 emissions; and is a major focus for emerging policies and targets at local to global scales (such as EUDR). To prevent and even reverse forest loss, we need to understand the underlying drivers of forest loss, and how these drivers vary geographically, and have been varying through time. Methodologically, this is an example of the rapidly advancing field of AI-powered remote sensing, using scaled up deep learning (specifically, specially adapted vision models) that can cope with the immense volume, complexity, and noise levels in satellite data and associated geospatial data. This work required advanced data processing, model and infrastructure development, expert human labelling, rigorous evaluation, domain expertise, and of course, great people, great team work … and time! Some examples of uses and potential impact: 🌲 Global Forest Watch leverages the drivers data as a key input for their global forest carbon flux model, enabling more accurate estimations of emission factors and, when combined with carbon flux data, identifying GHG emissions by specific drivers. 🌳 The Joint Research Centre (JRC) integrates the drivers of forest loss data into their 2020 global forest cover map, directly supporting the EU's regulation on deforestation-free supply chains. 🌴 The drivers data are already being used by WRI in methods being developed to account for land use change emissions in greenhouse gas inventories. 🍁 More generally, these detailed maps provide crucial insights into deforestation patterns and drivers, empowering local communities, policymakers, land managers, researchers and others to intervene effectively to prevent deforestation. To learn more, see the links below, which I stole without shame from a complementary LinkedIn post from our amazing WRI colleague Michelle Sims (hello and thanks Michelle!). And a shout out to some of the other key people involved:  Radost Stanimirova, PhD, Anton Raichuk, Maxim Neumann, Jessica Richter, Forrest Follett, @James MacCarthy, Kristine Lister, Christopher Randle, Lindsey Sloat, Elena Esipova, Jaelah Jupiter, Charlotte Y. Stanton, PhD, Dan Morris, Christy Melhart Slay, Nancy Harris. 👉 The ERL paper: https://lnkd.in/gxV34-3W 👉 Summary of key findings: https://lnkd.in/gghcyVzx 👉Technical blog: https://lnkd.in/gBv5ErKU Data is available on: 🌎 Google Earth Engine: https://lnkd.in/gt4t9zhp 🌏 World Resources Institute's Data Explorer: https://lnkd.in/gbVMUzpx 🌍 Global Forest Watch: https://gfw.global/2LUOmIx 🌍 Zenodo (including training + val data): https://lnkd.in/gsn-J9gg

Assessing urban tree cover using Sentinel-2A satellite data 🌳🛰️ #GreenCities #RemoteSensing #UrbanForests Tracking land conversion with Sentinel satellite data (2020-2024) 🛰️🌍 Remote sensing offers a powerful, accurate, and large-scale view of environmental changes, helping us understand urban growth, deforestation, and ecosystem shifts. The insights are crucial for sustainable planning and conservation! Step by step: 1️⃣ Data Acquisition 🛰️ • Collect Sentinel-2A/B satellite imagery from ESA’s Copernicus Open Access Hub. • Select images covering the study area from 2020 to 2024. • Apply cloud masking techniques to remove atmospheric disturbances. 2️⃣ Preprocessing 🔍 • Convert raw satellite images to reflectance values. • Apply geometric and radiometric corrections. • Use NDVI, NDBI, and other indices to enhance land cover features. 3️⃣ Land Cover Classification 🏙️🌿 • Apply machine learning (e.g., Random Forest, SVM) or deep learning models for classification. • Use supervised classification techniques with training datasets. • Categorize land types (urban, forest, water, agriculture, barren land). 4️⃣ Change Detection Analysis 🔄 • Compare classified images from different years (2020 vs. 2024). • Identify areas of deforestation, urban expansion, and land degradation. • Quantify changes in land cover percentages. 5️⃣ Validation & Accuracy Assessment ✅ • Use ground truth data and validation points. • Calculate accuracy metrics (confusion matrix, Kappa coefficient). • Refine classification model if needed. 6️⃣ Mapping & Visualization 🗺️ • Generate thematic maps for stakeholders. • Use GIS platforms (QGIS, ArcGIS) to present results. • Create interactive dashboards for decision-makers. 7️⃣ Insights & Decision Support 📊 • Provide reports and analytics on land conversion trends. • Offer recommendations for sustainable land management. • Support urban planning, forestry, and conservation efforts. 🔹 Results Impact: The analysis helps governments, businesses, and environmental agencies make data-driven decisions for sustainable development and resource management. #RemoteSensing #LandUseChange #GISMapping #SatelliteData #UrbanPlanning #EnvironmentalMonitoring #SustainableDevelopment #GeospatialAnalysis #EarthObservation #ClimateAction #Geolux #MappingSolutions #LandManagement #EcosystemMonitoring #RemoteSensing #LandUseChange #GISMapping #SatelliteData #EnvironmentalMonitoring #UrbanPlanning #Forestry #Deforestation #SmartCities #SustainableDevelopment #GeospatialAnalysis #EarthObservation #ClimateAction #PrecisionAgriculture #Geolux #MappingSolutions #LandManagement #EcosystemMonitoring #InfrastructurePlanning