Seven Swedish authorities co-operate to produce and maintain a new national land cover database. The project uses satellite data from the Sentinel-1 and Sentinel-2 satellites and national databases. Methods were developed in an ESA-financed project, CadasterENV Sweden. Based on the promising results from CadasterENV and the identified needs from many agencies and other users the Swedish Environmental Agency took the lead for a national co-operation with the goal to produce a national land cover database (NMD) and form an agreement for a continues maintenance of the information. The initial production phase is ready and published as open data. The steering group has agreed that an updated version should be published every 5th year. A lot of use cases are identified and to some extent already implemented e.g. statistical calculations, analyses of ecosystem services, green infrastructure and fire risks.

For 2019 the work consortium have been granted financing throw VINNOVA, Sweden’s Innovation Agency, programme “Challenge-driven innovation”. Challenge-driven innovation is an initiative that aims to solve social challenges that require broad cooperation to overcome. The solutions developed under the programme must make a clear contribution to one or more of the 17 Sustainable development goals laid out in the UN’s 2030 Agenda for sustainable development. Projects operating within the programme can tackle everything from climate change and health challenges to problems of inequality and the uneven distribution of resources.

The NMD project will ensure that there are continuously updated, high-quality data of the landscape and land-use in Sweden, on which there is a consensus among the relevant public agencies. There are several Agenda 2030 goals linked to the project; in particular Goal 15 Life on Land as it will contribute to the sustainable use of the ecosystem and its biological diversity. Land cover information can be used as a basis for environmental, societal and emergency planning. This helps to optimize the management of financial resources to counteract or slow down negative environmental impacts, including effects of climate change.

Currently, Swedish public agencies and municipalities make decisions that affect not only society as a whole but also individuals. These decisions can be based on insufficient, out-of-date and sometimes incorrect information. This lack of data, e.g. statistical monitoring of the ecosystem and biological diversity is even worse on a national level. Additionally, the results of the processes that need to take environmental aspects into account are worsened. The socio-economic aspects are also affected, e.g. when incomplete data is used in an early stage of an environmental assessment, which can lead to a rejection of exploitation plans at a much later date. Future climate change will also create increased pressure on, as well as objections to, different types of land use including forestry and agriculture. In order to make the right decisions based on the current situation, predicative scenarios need to have complete and updated information.

The problem with insufficient support material affects not only governmental agencies but also private individuals working on social causes. In the long run, this will even affect individuals who have to take the decisions made by public agencies into account or face the consequences of actions that negatively affect the environment and its resilience, as well as human health. In order to meet this societal challenge, there needs to be not only must a common and accepted vision of Sweden’s landscapes but also an active management and update of this vision. To succeed in this endeavor, an interdepartmental infrastructure for effectively exchanging geographical information will need to be created. There is a clear need for a technical development plan for the long-term and properly managed distribution of land cover data. These maps need to be kept current. Therefore, new approaches and innovative methods for continually and cost-effectively updating the data are needed. Such updating and management would also give an opportunity to improve and expand the mapping to provide more information than what is possible today in order to meet the needs of the future.

The challenge will be to develop innovative methods that take advantage of the latest technology (including automation, time series analysis and machine learning) allowing for efficient updating of the NMD. Additionally, there is a need to further develop the thematic content and quality of the NMD. The amount of data, comprised of satellite images, 3D laser point clouds, cadastral information and other pertinent data, is large and constantly increasing, requiring innovative data management solutions. Consequently, the coordination of geographic data from a large number of different agencies is an important part of the project, relating to the Swedish government’s Digital First initiative.

The Swedish authorities have conducted a workshop to identify SDGs an up-to-date Land Cover data has a potential to contribute. All the main goals and their targets were included in the exercise. The result of the workshop showed that for goal 15, 75% of the targets are estimated to benefit from NMD. Goals 11 and 13 also stand out, where 70% and 60% of the targets are considered to benefit.

The presentation will focus both on the national co-operation related to the project performance, communication with user groups and the challenging work to set up the structure and co-operation for a sustainable maintenance of the database. Practical examples of how NMD will contribute to the SDGs will be demonstrated.

More than any other sector, agriculture lies at the heart of the 2030 agenda. By investing in agriculture, we do not only address hunger and malnutrition, but also sustainable use of land and water resources, poverty, and climate change. Agriculture has also been one of the first areas where Earth Observation has played an important role. Earth Observation services can contribute to increased food security and a more sustainable and resilient agricultural system.

Progress in SDG2 (zero hunger) is highly dependent on progress in ensuring sustainable management of water and land. One of the means to “increase agricultural productivity” (target 2.4) is irrigation. Agriculture is the major water user and “substantially increasing water use efficiency” (target 6.4) is key. Agriculture is also one of the key drivers of land use change, land degradation and desertification, and increased agricultural productivity should be achieved without adverse impacts on land, water and biodiversity and without further deforestation, biodiversity losses and climate change impact. Target 15.4 calls for sustainable use of land resources to “combat desertification and achieve a land degradation neutral world”.

Earth Observation can play a significant role in providing “high-quality, timely and reliable data” (target 17.18) as it provides the synoptic view of large areas at a time in a consistent, reliable and cost-effective manner. The European Space Agency (ESA) launched the Earth Observation for Sustainable Development (EO4SD) initiative to demonstrate the Multilateral Development Banks (MDB’s) and their Client States that Earth Observation services measurably enhance the effectiveness of their investments in sustainable development.

This paper highlights demonstrations of the EO4SD initiative’s agriculture cluster activity in mainstreaming satellite Earth Observation services in agricultural development projects in Africa, Asia and Latin America. EO4SD activities support agricultural development projects by monitoring progress and evaluating impact using project indicators closely related to the SDG indicators focussing on increased agricultural productivity, improved agricultural water management, and fostering land degradation neutrality.

Services include agricultural productivity assessments, crop area and type estimates, irrigation water use efficiency, sustainable land management, and environmental impact assessments of agriculture.
• Agricultural productivity is the output in relation to resources, such as land, labour, capital, and water. Earth Observation can measure the distribution, status and dynamics of agricultural production with products related to land utilization (cultivated/irrigated area, crop type), cropping intensity (phenology), and crop production (e.g. proxies for yield such as vegetation cover and biomass production).
• Agricultural water management: Satellite data allows the monitoring of droughts and floods to support agricultural risk management plans. Earth Observation supports sustainable agricultural water use and promotes sustainable agricultural technologies by mapping water consumption, water stress and water productivity, which are measures of irrigation performance and water use efficiency.
• Enable sustainable land management: Land cover and vegetation dynamics maps derived from Earth Observation support sustainable land management by showing how agriculture impacts on land degradation, identifying focus areas (regreening and degradation), and monitoring changes in land cover, e.g. deforestation.

Sustainable Development Goals (SDG) indicator 6.4.1 calls for monitoring change in water use efficiency over time with the aim of assessing the influence of economic development on the utilization of fresh water resources. It focuses on three major economic domains: agriculture; industry and mining; and other services. While the use of satellite observations for wholistic monitoring of water use efficiency in the services and industrial sectors is not a viable approach, monitoring the water use efficiency in the agricultural sector, and in particular in crop irrigation, is much more feasible. It has been estimated that in 2015 in Uganda 41% of fresh water withdrawals were used in agriculture and that 76% of irrigated areas were located in large-scale (>500 ha) irrigation schemes. Those numbers are expected to increase significantly in the coming years as the government is planning an expansion of the irrigation systems to cope with pressures of population growth, climate change and economic development. Therefore, reliable and independent monitoring of water use in this sector is of high importance for reporting of the SDG indicator 6.4.1 in Uganda.

Currently water use efficiency in Uganda in the agricultural sector is not monitored in a detailed manner and relies mostly on self-reporting of the water license holders. As part of the Earth Observation for Sustainable Development Goals project, we have developed a system utilizing observations from Sentinel-2 and Sentinel-3 satellites to monitor parcel-scale actual evapotranspiration (ET) of large irrigation schemes in Uganda and turn this information into data required for SDG 6.4.1 reporting. The system consists of three main modules. Firstly, national-scale mapping of irrigated areas is performed through the analysis of timeseries of Sentinel-2 images. Secondly, the actual ET of the previously identified areas is mapped at 20 m resolution by fusing optical information from Sentinel-2 satellites together with thermal information from Sentinel-3 satellites and by employing physical models of land surface energy fluxes. Finally, the actual amount of water used for irrigation is assessed by subtracting rainfall from the ET and this data is combined with crop yield estimates to derive economic output per unit of water used.

The developed system fully exploits the spatial, temporal and spectral properties of the Sentinel-2 and Sentinel-3 satellite constellation. It is not reliant on any local-gathered data (although it can be included for increased accuracy) and thus can provide an independent synoptic view of water use in the agricultural sector in countries with limited data collection infrastructure. Therefore, the use of the system can be extended beyond Uganda and beyond SDG reporting to, for example, monitoring of water license compliance.

An estimated 40% of the world’s crops are lost to pests, impacting on the ability of smallholder farmers to feed their families and compromising food security. These losses also affect international trade, food supply chains and hamper the pursuit of UN Sustainable Development Goals (SDGs) 1 and 2. Tuta absoluta (Tomato Leafminer) in West Africa and Spodoptera frugiperda (Fall Armyworm) in Western and Southern Africa have had devastating impacts at least in part due to the unexpected nature of the outbreaks and more information is needed on the environmental conditions that drive pest growth and distribution to help preparedness. PRISE (Pest Risk Information SErvice) is a 5-year project, funded by the UK Space Agency International Partnership programme (IPP), which integrates openly available Earth Observation (EO) and meteorological data into complex pest development models to provide Sub-Saharan African (SSA) farmers with advanced pest risk warnings to help minimise crop losses. This will be achieved by:

1. Using openly available EO and meteorological data sources to derive the environmental parameters required to model crop pest development, such as air and crop canopy temperature and rainfall.

2. Using these derived parameters to model the development of pests to understand pest risk in SSA.

3. Developing the PRISE Agro-Met Data Cube (AMDQ) to facilitate geospatial data access for risk analysis and to federate the effort of developing new models in the wider community.

4. Creating long-term climatologies of derived environmental parameters that allow current conditions to be compared with historic trends.

PRISE is a collaboration between Earth Observation scientists at Assimila Ltd and Kings College London, plant pathologists and social scientists at CABI, and in-country contributors from government and research organisations in the partner countries of Kenya, Ghana and Zambia

The aim of the PRISE project is to directly contribute to the UN SDGs 1, 2 and 12. Specifically, to addresses targets: 1.1, alleviating extreme poverty; 1.5, reducing vulnerability to climate related extreme events; 2.1, ensuring access by all people to safe, nutritious and sufficient food; 2.3, doubling the agricultural productivity and income of small-scale food producers; 2.4, implement resilience agricultural practices that increase agricultural productivity in the face of climate change and other disasters; and 2.5, increased investment in extension services and technology. SDG 12 Target 12.4 will also be addressed, achieving environmentally sound management of chemicals. All the above SDGs cannot be achieved without strong partnerships (SDG 17) between project partners and key stakeholders such as the Ministries of Agriculture, Metrological Departments and agricultural research agencies in the three partner countries.

Pest development is strongly linked to environmental conditions such as temperature, rainfall and humidity. It is possible to obtain this data from ground-based sensors, but coverage is often sparse and data collection is resource-intensive. EO products provide an efficient source of data in a well-calibrated, spatially continuous and reliable manner, which can be used to supplement ground-based observations.

There are, however, still challenges when with working with Earth Observation, which are manifest more greatly for users with limited technological resources. These challenges can be grouped into three broad categories: access, pre-processing and analysis. Data access for a range of environmental datasets includes traversing the many different source providers, each with their own authentication protocols and documentation on product use, as well as downloading data which can be slow on unreliable connections. Pre-processing EO data requires significant compute resource to generate analysis-ready data from the multiple datasets, each with varying file formats, resolutions and projections. Finally, the analysis of this data is also compute-intensive task and the filtering of data down to the specific time or region of interest can be a lengthy process, taking time away from the analysis itself.

The PRISE AMDQ provides a solution to these issues, for both UK and in-country partners. It removes the data download difficulties through the provision of tools that access multiple data portals to automatically download or access data in a cloud environment. The acquired data are then used to create seamless and homogenous analysis-ready datasets using user-specified resolutions, projections and formats. These datasets in turn support smart spatio-temporal querying using latitude, longitude and time in natural language, as well as traditional indexing using rows, columns and layers.

The AMDQ forms the technical foundation for the PRISE project. It automatically gathers EO data from a range of providers and connects seamlessly to the PRISE universal algorithm processing engine which automatically generates pest risk estimations. The AMDQ serves this information via an API to the dissemination partners who in turn pass pest development information to in-country experts. And last but not least, the AMDQ serves as an analysis-ready data access portal for all partners in PRISE, supporting the further development of downstream applications which reduce crop losses to pests in Ghana, Zambia and Kenya.

Progress towards sustainable forest management is the agreed indicator to monitor countries efforts to promote the implementation of sustainable management of all types of forest, halt deforestation, restore degraded forest and increase afforestation and reforestation by 2020 (SDG 15.2). Estimation of forested areas is insufficient to track advances towards the economic, social and environmental dimensions of Sustainable Forest Management (SFM), and hence five sub-indicators, integrated in a dashboard have been proposed to measure progress towards all dimensions of SFM. The UN Statistical Division recognises that the current sub-set of indicators do not fully cover all aspects of SFM, particularly the social and economic. Hence, this presentation investigates the potential of a new sub-indicator that uses remote sensing observations to assess the supply of carbon sequestration at ecosystem level, coupling the output with a model to compute the social cost of carbon.
Tropical Dry Forests (TDFs) in Mexico, Costa Rica, Brazil and Bolivia are the forest ecosystem selected to assess their potential as carbon sequestration pools, and the associated environmental and socio-economic values. The methodological framework includes biophysical quantification of TDF’s carbon sequestration using a remotely sensed spatially explicit approach to map the relationship between productivity and its equivalent to total carbon dioxide (CO2) sequestration per spatial unit in each of the sites. MODIS-derived annual trends from 2000 to 2015 were considered to that end. Carbon use efficiency was computed to analyse the capacity of this ecosystem to sequester CO2, and as input to economic valuation for inferring the monetary cost of carbon sequestration in TDFs. Three reference values of the Social Cost of Carbon were applied and compared.
Costa Rica is the site with the highest social cost of carbon over the period 2000-2015 (USD 772 to 4468 per hectare, depending on the reference value applied); Bolivia ranks last at USD 237 to 1370 per ha. Results show that TDFs of the Americas supply an annual average CO2 sequestration rate of 22.34 ± 3.31 tCO2 ha-1 yr-1, which equates to an economic value ranging from $488.8 USD ha-1 yr-1, for the lowest estimate, to $2828.5 USD ha-1 yr-1 for the highest value. Extrapolating our results to prior estimations of the extent of TDFs in the Americas evidence that his ecosystem has a potential of sequestering approximately 1.16 gigatons of CO2 per year, with an equivalent economic value that ranges between USD 25 billion for the lowest estimate to USD 146 billion for the highest estimate.
Furthermore, these results show that using spatially explicit remote sensing based approaches to estimate the capacity of specific forest ecosystems for sequestrating CO2, and coupling the output to conventional models of carbon costing offers a viable alternative for countries to better estimate the social and economic aspects of SFM. Given the high temporal resolution of sensors such as MODIS, Sentinel, accounting for ‘variation over time and between countries’ of sustainable forest management becomes a reality.

Public urban spaces such as streets, open spaces, green areas, parks, and public buildings are a big part of cities that are often overlooked. Inadequate, poorly designed, or privatized public spaces often generate exclusion and marginalization and degrades the livability of the urban environment. The importance of public space is recently highlighted in international agendas (e.g. SDG11.7) and diverse organizations started piloting the role of urban green and public spaces planning in cities. While a systematic classification is needed for cities to quantify and manage their assets and policies efficiently, it is surprisingly rare, especially in developing countries.

Even though definitions of public spaces have been actively discussed and evolved over decades, there is still no universally agreed-upon definition. This is mainly because the meaning of “public” matters. When it comes to the definition of public urban space, there are three categories of physical types, as proposed by UN-Habitat: streets, open green spaces, and public buildings, but the question remains how to classify a public space in a more specific way.
Therefore, the World Bank Group’ UrbanScapes multidisciplinary team has partnered with the European Space Agency (ESA) under its Earth Observation for Sustainable Development (EO4SD) initiative’s dedicated activity cluster on urban development. GISAT, one of the consortium partners, provides the team with support on developing a rule-based identification of public spaces by determining their physical characteristics using Earth Observation techniques, focusing on six criteria - distance to roads, proximity to water, compactness, land cover patterns, location and size - to came up with 12 classes of public spaces as piloted in Dhaka, Bangladesh (see https://blogs.worldbank.org/sustainablecities/quantifying-public-spaces-better-quality-urban-assets) and Karachi, Pakistan.

As part of this program, the UrbanScapes Group cooperated with GISAT to use EO based classification of urban public spaces to better prioritize their intervention areas during the project's preparation. The presentation will provide an overview on a spatial analytical work done, the way how it was applied to the Bank operations as well as the potential for streamlining the approach for future Bank operations discussing the robustness, standardization and transferability aspects linked to other cities (Bamako, Lima, Falluja and Ramadi) application.

Earth Observation data potential exploration part is carried out by cooperation in the frame of the EO4SD-Urban project supported by the European Spatial Agency. EO4SD (Earth Observation for Sustainable Development) is a new ESA initiative, built on previous successful ESA-WBG collaborations, aims to achieve a step increase in the uptake of satellite-based environmental information in the IFIs regional and global programs. It follows a systematic, user-driven approach in order to meet longer-term, strategic geospatial information needs in the individual developing countries, as well as international and regional development organizations (http://eo4sd.esa.int).