DeforestationDeforestation or forest clearance is the removal of a forest or stand of trees from land that is then converted to non-forest use. Deforestation can involve conversion of forest land to farms, ranches, or urban use. The most concentrated deforestation occurs in tropical rainforests. About 31% of Earth's land surface is covered by forests at present. This is one-third less than the forest cover before the expansion of agriculture, with half of that loss occurring in the last century.
Deforestation and climate changeDeforestation is a primary contributor to climate change, and climate change affects forests. Land use changes, especially in the form of deforestation, are the second largest anthropogenic source of atmospheric carbon dioxide emissions, after fossil fuel combustion. Greenhouse gases are emitted during combustion of forest biomass and decomposition of remaining plant material and soil carbon. Global models and national greenhouse gas inventories give similar results for deforestation emissions.
Tropical rainforestTropical rainforests are rainforests that occur in areas of tropical rainforest climate in which there is no dry season – all months have an average precipitation of at least 60 mm – and may also be referred to as lowland equatorial evergreen rainforest. True rainforests are typically found between 10 degrees north and south of the equator (see map); they are a sub-set of the tropical forest biome that occurs roughly within the 28-degree latitudes (in the equatorial zone between the Tropic of Cancer and Tropic of Capricorn).
Machine learningMachine learning (ML) is an umbrella term for solving problems for which development of algorithms by human programmers would be cost-prohibitive, and instead the problems are solved by helping machines 'discover' their 'own' algorithms, without needing to be explicitly told what to do by any human-developed algorithms. Recently, generative artificial neural networks have been able to surpass results of many previous approaches.
Greenhouse gas emissionsGreenhouse gas emissions (abbreviated as GHG emissions) from human activities strengthen the greenhouse effect, contributing to climate change. Carbon dioxide (), from burning fossil fuels such as coal, oil, and natural gas, is one of the most important factors in causing climate change. The largest emitters are China followed by the US, although the United States has higher emissions per capita. The main producers fueling the emissions globally are large oil and gas companies.
Carbon emission tradingEmission trading (ETS) for carbon dioxide (CO2) and other greenhouse gases (GHG) is a form of carbon pricing; also known as cap and trade (CAT) or carbon pricing. It is an approach to limit climate change by creating a market with limited allowances for emissions. This can lower competitiveness of fossil fuels and accelerate investments into low carbon sources of energy such as wind power and photovoltaics. Fossil fuels are the main driver for climate change. They account for 89% of all CO2 emissions and 68% of all GHG emissions.
Attention (machine learning)Machine learning-based attention is a mechanism mimicking cognitive attention. It calculates "soft" weights for each word, more precisely for its embedding, in the context window. It can do it either in parallel (such as in transformers) or sequentially (such as recursive neural networks). "Soft" weights can change during each runtime, in contrast to "hard" weights, which are (pre-)trained and fine-tuned and remain frozen afterwards. Multiple attention heads are used in transformer-based large language models.
Satellite imagerySatellite images (also Earth observation imagery, spaceborne photography, or simply satellite photo) are s of Earth collected by imaging satellites operated by governments and businesses around the world. Satellite imaging companies sell images by licensing them to governments and businesses such as Apple Maps and Google Maps. The first images from space were taken on sub-orbital flights. The U.S-launched V-2 flight on October 24, 1946, took one image every 1.5 seconds.
Deep learningDeep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the use of multiple layers in the network. Methods used can be either supervised, semi-supervised or unsupervised.
Embedded emissionsOne way of attributing greenhouse gas (GHG) emissions is to measure the embedded emissions of goods that are being consumed (also referred to as "embodied emissions", "embodied carbon emissions", or "embodied carbon"). This is different from the question of to what extent the policies of one country to reduce emissions affect emissions in other countries (the "spillover effect" and "carbon leakage" of an emissions reduction policy). The UNFCCC measures emissions according to production, rather than consumption.
Emission intensityLife-cycle greenhouse gas emissions of energy sources An emission intensity (also carbon intensity or C.I.) is the emission rate of a given pollutant relative to the intensity of a specific activity, or an industrial production process; for example grams of carbon dioxide released per megajoule of energy produced, or the ratio of greenhouse gas emissions produced to gross domestic product (GDP).
Carbon offsets and creditsA carbon offset is a reduction or removal of emissions of carbon dioxide or other greenhouse gases made in order to compensate for emissions made elsewhere. A carbon credit or offset credit is a transferrable financial instrument (i.e. a derivative of an underlying commodity) certified by governments or independent certification bodies to represent an emission reduction that can then be bought or sold. Both offsets and credits are measured in tonnes of carbon dioxide-equivalent (CO2e).
Carbon accountingCarbon accounting (or greenhouse gas accounting) is a framework of methods to measure and track how much greenhouse gas (GHG) an organization emits. It can also be used to track projects or actions to reduce emissions in sectors such as forestry or renewable energy. Corporations, cities and other groups use these techniques to help limit climate change. Organizations will often set an emissions baseline, create targets for reducing emissions, and track progress towards them.
Biodiversity lossBiodiversity loss includes the worldwide extinction of different species, as well as the local reduction or loss of species in a certain habitat, resulting in a loss of biological diversity. The latter phenomenon can be temporary or permanent, depending on whether the environmental degradation that leads to the loss is reversible through ecological restoration/ecological resilience or effectively permanent (e.g. through land loss).
BiodiversityBiodiversity or biological diversity is the variety and variability of life on Earth. Biodiversity is a measure of variation at the genetic (genetic variability), species (species diversity), and ecosystem (ecosystem diversity) level. Biodiversity is not distributed evenly on Earth; it is usually greater in the tropics as a result of the warm climate and high primary productivity in the region near the equator. Tropical forest ecosystems cover less than 10% of earth's surface and contain about 90% of the world's species.
REDD and REDD+REDD originally referred to "reducing emissions from deforestation in developing countries", which was the title of the original document on REDD. It was superseded by REDD+ in the Warsaw Framework on REDD-plus negotiations. REDD+ (or REDD-plus) refers to "reducing emissions from deforestation and forest degradation in developing countries, and the role of conservation, sustainable management of forests, and enhancement of forest carbon stocks in developing countries" (emphasis added).
Carbon budgetA carbon budget is a concept used in climate policy to help set emissions reduction targets in a fair and effective way. It looks at "the maximum amount of cumulative net global anthropogenic carbon dioxide () emissions that would result in limiting global warming to a given level". When expressed relative to the pre-industrial period it is referred to as the total carbon budget, and when expressed from a recent specified date it is referred to as the remaining carbon budget.
Deep reinforcement learningDeep reinforcement learning (deep RL) is a subfield of machine learning that combines reinforcement learning (RL) and deep learning. RL considers the problem of a computational agent learning to make decisions by trial and error. Deep RL incorporates deep learning into the solution, allowing agents to make decisions from unstructured input data without manual engineering of the state space. Deep RL algorithms are able to take in very large inputs (e.g.
Biodiversity bankingBiodiversity banking, also known as biodiversity trading or conservation banking, biodiversity mitigation banks, compensatory habitat, set-asides, biodiversity offsets, are conservation activities that compensate for the loss of biodiversity with the goal of biodiversity maintenance through a framework which allows biodiversity to be reliably measured, and market based solutions applied to improving biodiversity. Biodiversity banking provides a means to place a monetary value on ecosystem services.
Cognitive architectureA cognitive architecture refers to both a theory about the structure of the human mind and to a computational instantiation of such a theory used in the fields of artificial intelligence (AI) and computational cognitive science. The formalized models can be used to further refine a comprehensive theory of cognition and as a useful artificial intelligence program. Successful cognitive architectures include ACT-R (Adaptive Control of Thought - Rational) and SOAR.
