A Unified Cropland Layer at 250 m for Global Agriculture Monitoring

Satellite mapping reveals extensive industrial activity at sea

The world's population increasingly relies on the ocean for food, energy production and global trade1-3, yet human activities at sea are not well quantified4,5. We combine satellite imagery, vessel GPS data and deep-learning models to map industrial vessel activities and offshore energy infrastructure across the world's coastal waters from 2017 to 2021. We find that 72-76% of the world's industrial fishing vessels are not publicly tracked, with much of that fishing taking place around South Asia, Southeast Asia and Africa. We also find that 21-30% of transport and energy vessel activity is missing from public tracking systems. Globally, fishing decreased by 12 ± 1% at the onset of the COVID-19 pandemic in 2020 and had not recovered to pre-pandemic levels by 2021. By contrast, transport and energy vessel activities were relatively unaffected during the same period. Offshore wind is growing rapidly, with most wind turbines confined to small areas of the ocean but surpassing the number of oil structures in 2021. Our map of ocean industrialization reveals changes in some of the most extensive and economically important human activities at sea.

Crop Type Maps for Operational Global Agricultural Monitoring

Crop type maps identify the spatial distribution of crop types and underpin a large range of agricultural monitoring applications ranging from early warning of crop shortfalls, crop condition assessments, production forecasts, and damage assessment from extreme weather, to agricultural statistics, agricultural insurance, and climate mitigation and adaptation decisions. Despite their importance, harmonized, up-to-date global crop type maps of the main food commodities do not exist to date. To address this critical data gap of global-scale consistent, up-to-date crop type maps, we harmonized 24 national and regional datasets from 21 sources covering 66 countries to develop a set of Best Available Crop Specific masks (BACS) over the major production and export countries for wheat, maize, rice, and soybeans, in the context of the G20 Global Agriculture Monitoring Program, GEOGLAM.

Analysing the relationship between human modification and land surface temperature fluctuation in the Ramganga basin, India

In many regions across the world, including river basins, population growth and land development have enhanced the demand for land and other natural resources. The anthropogenic activities can be detrimental to the vital ecosystems that sustain the river basin region. This work assessed the impact of human modification on land surface temperature (LST) for the Ramganga basin in India. It has been hypothesised that the footprints of anthropogenic activities in the region have been connected to the LST fluctuation for the region, which could indicate environmental degradation. The LST variation between 2000 and 2016 has been estimated to test this hypothesis. The spatio-temporal correlation between human modification and LST has been computed. LST has been calculated with MODIS satellite data in the Google earth engine (GEE) platform, and anthropogenic activities can be visualised using an LU/LC map of the basin created by the Classification and Regression (CART) technique. The statistical parameters (average, maximum and standard deviation) of annual temperature for each pixel in 17 years (2000-2016) have been assessed to establish the links with human modification. The result of this work portrays a positive correlation of 0.705 between maximum LST and human modification. The forest class in the basin region has the lowest average human modification value (0.37), and it also possesses the lowest mean LST of 26.72 °C. Similarly, the settlement class has the highest average human modification value (0.85), and the mean LST temperature of this class has been on the higher side, having a value of 31.07 °C.

National-Scale Cropland Mapping Based on Phenological Metrics, Environmental Covariates, and Machine Learning on Google Earth Engine

Many challenges prevail in cropland mapping over large areas, including dealing with massive volumes of datasets and computing capabilities. Accordingly, new opportunities have been opened at a breakneck pace with the launch of new satellites, the continuous improvements in data retrieval technology, and the upsurge of cloud computing solutions such as Google Earth Engine (GEE). Therefore, the present work is an attempt to automate the extraction of multi-year (2016–2020) cropland phenological metrics on GEE and use them as inputs with environmental covariates in a trained machine-learning model to generate high-resolution cropland and crop field-probabilities maps in Morocco. The comparison of our phenological retrievals against the MODIS phenology product shows very close agreement, implying that the suggested approach accurately captures crop phenology dynamics, which allows better cropland classification. The entire country is mapped using a large volume of reference samples collected and labelled with a visual interpretation of high-resolution imagery on Collect-Earth-Online, an online platform for systematically collecting geospatial data. The cropland classification product for the nominal year 2019–2020 showed an overall accuracy of 97.86% with a Kappa of 0.95. When compared to Morocco’s utilized agricultural land (SAU) areas, the cropland probabilities maps demonstrated the ability to accurately estimate sub-national SAU areas with an R-value of 0.9. Furthermore, analyzing cropland dynamics reveals a dramatic decrease in the 2019–2020 season by 2% since the 2018–2019 season and by 5% between 2016 and 2020, which is partly driven by climate conditions, but even more so by the novel coronavirus disease 2019 (COVID-19) that impacted the planting and managing of crops due to government measures taken at the national level, like complete lockdown. Such a result proves how much these methods and associated maps are critical for scientific studies and decision-making related to food security and agriculture.

Usage of Artificial Intelligence and Remote Sensing as Efficient Devices to Increase Agricultural System Yields

Artificial Intelligence is an emerging technology in the field of agriculture. Artificial Intelligence-based tools and equipment have actually taken the agriculture sector to a different level. This new technology has improved crop production and enhanced instantaneous monitoring, processing, and collection. The most recent computerized structures using remote sensing and drones have made a significant contribution to the agro-based domain. Moreover, remote sensing has the capability to support the development of farming applications with the aim of facing this main defy, via giving cyclic records on yield status during studied periods at diverse degrees and for diverse parameters. Various hi-tech, computer-supported structures are created to determine different central factors such as plant detection, yield recognition, crop quality, and several other methods. This paper includes the techniques employed for the analysis of collected information in order to enhance the productivity, forecast eventual threats, and reduce the task load on cultivators.

Multiple cropping systems of the world and the potential for increasing cropping intensity

Multiple cropping, defined as harvesting more than once a year, is a widespread land management strategy in tropical and subtropical agriculture. It is a way of intensifying agricultural production and diversifying the crop mix for economic and environmental benefits. Here we present the first global gridded data set of multiple cropping systems and quantify the physical area of more than 200 systems, the global multiple cropping area and the potential for increasing cropping intensity. We use national and sub-national data on monthly crop-specific growing areas around the year 2000 (1998-2002) for 26 crop groups, global cropland extent and crop harvested areas to identify sequential cropping systems of two or three crops with non-overlapping growing seasons. We find multiple cropping systems on 135 million hectares (12% of global cropland) with 85 million hectares in irrigated agriculture. 34%, 13% and 10% of the rice, wheat and maize area, respectively are under multiple cropping, demonstrating the importance of such cropping systems for cereal production. Harvesting currently single cropped areas a second time could increase global harvested areas by 87-395 million hectares, which is about 45% lower than previous estimates. Some scenarios of intensification indicate that it could be enough land to avoid expanding physical cropland into other land uses but attainable intensification will depend on the local context and the crop yields attainable in the second cycle and its related environmental costs.

Comparison of Different Cropland Classification Methods under Diversified Agroecological Conditions in the Zambezi River Basin

Having updated knowledge of cropland extent is essential for crop monitoring and food security early warning. Previous research has proposed different methods and adopted various datasets for mapping cropland areas at regional to global scales. However, most approaches did not consider the characteristics of farming systems and apply the same classification method in different agroecological zones (AEZs). Furthermore, the acquisition of in situ samples for classification training remains challenging. To address these knowledge gaps and challenges, this study applied a zone-specific classification by comparing four classifiers (random forest, the support vector machine (SVM), the classification and regression tree (CART) and minimum distance) for cropland mapping over four different AEZs in the Zambezi River basin (ZRB). Landsat-8 and Sentinel-2 data and derived indices were used and synthesized to generate thirty-five layers for classification on the Google Earth Engine platform. Training samples were derived from three existing landcover datasets to minimize the cost of sample acquisitions over the large area. The final cropland map was generated at a 10 m resolution. The performance of the four classifiers and the viability of training samples were analysed. All classifiers presented higher accuracy in cool AEZs than in warm AEZs, which may be attributed to field size and lower confusion between cropland and grassland classes. This indicates that agricultural landscape may impact classification results regardless of the classifiers. Random forest was found to be the most stable and accurate classifier across different agricultural systems, with an overall accuracy of 84% and a kappa coefficient of 0.67. Samples extracted over the full agreement areas among existing datasets reduced uncertainty and provided reliable calibration sets as a replacement of costly in situ measurements. The methodology proposed by this study can be used to generate periodical high-resolution cropland maps in ZRB, which is helpful for the analysis of cropland extension and abandonment as well as intensity changes in response to the escalating population and food insecurity.

Predicting Soybean Yield at the Regional Scale Using Remote Sensing and Climatic Data

Crop yield modeling at the regional level is one of the most important methods to ensure the profitability of the agro-industrial economy and the solving of the food security problem. Due to a lack of information about crop distribution over large agricultural areas, as well as the crop separation problem (based on remote sensing data) caused by the similarity of phenological cycles, a question arises regarding the relevance of using data obtained from the arable land mask of the region to predict the yield of individual crops. This study aimed to develop a regression model for soybean crop yield monitoring in municipalities and was conducted in the Khabarovsk Territory, located in the Russian Far East. Moderate Resolution Imaging Spectroradiometer (MODIS) data, an arable land mask, the meteorological characteristics obtained using the VEGA-Science web service, and crop yield data for 2010–2019 were used. The structure of crop distribution in the Khabarovsk District was reproduced in experimental fields, and Normalized Difference Vegetation Index (NDVI) seasonal variation approximating functions were constructed (both for total district sown area and different crops). It was found that the approximating function graph for the experimental fields corresponds to a similar graph for arable land. The maximum NDVI forecast error on the 30th week in 2019 using the approximation parameters according to 2014–2018 did not exceed 0.5%. The root-mean-square error (RMSE) was 0.054. The maximum value of the NDVI, as well as the indicators characterizing the temperature regime, soil moisture, and photosynthetically active radiation in the region during the period from the 1st to the 30th calendar weeks of the year, were previously considered as parameters of the regression model for predicting soybean yield. As a result of the experiments, the NDVI and the duration of the growing season were included in the regression model as independent variables. According to 2010–2018, the mean absolute percentage error (MAPE) of the regression model was 6.2%, and the soybean yield prediction absolute percentage error (APE) for 2019 was 6.3%, while RMSE was 0.13 t/ha. This approach was evaluated with a leave-one-year-out cross-validation procedure. When the calculated maximum NDVI value was used in the regression equation for early forecasting, MAPE in the 28th–30th weeks was less than 10%.

A Data-Intensive Approach to Address Food Sustainability: Integrating Optic and Microwave Satellite Imagery for Developing Long-Term Global Cropping Intensity and Sowing Month from 2001 to 2015

It is necessary to develop a sustainable food production system to ensure future food security around the globe. Cropping intensity and sowing month are two essential parameters for analyzing the food–water–climate tradeoff as food sustainability indicators. This study presents a global-scale analysis of cropping intensity and sowing month from 2000 to 2015, divided into three groups of years. The study methodology integrates the satellite-derived normalized vegetation index (NDVI) of 16-day composite Moderate Resolution Imaging Spectroradiometer (MODIS) and daily land-surface-water coverage (LSWC) data obtained from The Advanced Microwave Scanning Radiometer (AMSR-E/2) in 1-km aggregate pixel resolution. A fast Fourier transform was applied to normalize the MODIS NDVI time-series data. By using advanced methods with intensive optic and microwave time-series data, this study set out to anticipate potential dynamic changes in global cropland activity over 15 years representing the Millennium Development Goal period. These products are the first global datasets that provide information on crop activities in 15-year data derived from optic and microwave satellite data. The results show that in 2000–2005, the total global double-crop intensity was 7.1 million km2, which increased to 8.3 million km2 in 2006–2010, and then to approximately 8.6 million km2 in 2011–2015. In the same periods, global triple-crop agriculture showed a rapid positive growth from 0.73 to 1.12 and then 1.28 million km2, respectively. The results show that Asia dominated double- and triple-crop growth, while showcasing the expansion of single-cropping area in Africa. The finer spatial resolution, combined with a long-term global analysis, means that this methodology has the potential to be applied in several sustainability studies, from global- to local-level perspectives.

Assessing the Link between Human Modification and Changes in Land Surface Temperature in Hainan, China Using Image Archives from Google Earth Engine

In many areas of the world, population growth and land development have increased demand for land and other natural resources. Coastal areas are particularly susceptible since they are conducive for marine transportation, energy production, aquaculture, marine tourism and other activities. Anthropogenic activities in the coastal areas have triggered unprecedented land use change, depletion of coastal wetlands, loss of biodiversity, and degradation of other vital ecosystem services. The changes can be particularly drastic for small coastal islands with rich biodiversity. In this study, the influence of human modification on land surface temperature (LST) for the coastal island Hainan in Southern China was investigated. We hypothesize that for this island, footprints of human activities are linked to the variation of land surface temperature, which could indicate environmental degradation. To test this hypothesis, we estimated LST changes between 2000 and 2016 and computed the spatio-temporal correlation between LST and human modification. Specifically, we classified temperature data for the four years 2000, 2006, 2012 and 2016 into 5 temperature zones based on their respective mean and standard deviation values. We then assessed the correlation between each temperature zone and a human modification index computed for the year 2016. Apart from this, we estimated mean, maximum and the standard deviation of annual temperature for each pixel in the 17 years to assess the links with human modification. The results showed that: (1) The mean LST temperature in Hainan Island increased with fluctuations from 2000 to 2016. (2) The moderate temperature zones were dominant in the island during the four years included in this study. (3) A strong positive correlation of 0.72 between human modification index and mean and maximum LST temperature indicated a potential link between human modification and mean and maximum LST temperatures over the 17 years of analysis. (4) The mean value of human modification index in the temperature zones in 2016 showed a progressive rise with 0.24 in the low temperature zone, 0.33 in the secondary moderate, 0.45 in the moderate, 0.54 in the secondary high and 0.61 in the high temperature zones. This work highlighted the potential value of using large and multi-temporal earth observation datasets from cloud platforms to assess the influence of human activities in sensitive ecosystems. The results could contribute to the development of sustainable management and coastal ecosystems conservation plans.

How Response Designs and Class Proportions Affect the Accuracy of Validation Data

Reference data collected to validate land-cover maps are generally considered free of errors. In practice, however, they contain errors despite best efforts to minimize them. These errors propagate during accuracy assessment and tweak the validation results. For photo-interpreted reference data, the two most widely studied sources of error are systematic incorrect labeling and vigilance drops. How estimation errors, i.e., errors intrinsic to the response design, affect the accuracy of reference data is far less understood. In this paper, we analyzed the impact of estimation errors for two types of classification systems (binary and multiclass) as well as for two common response designs (point-based and partition-based) with a range of sub-sample sizes. Our quantitative results indicate that labeling errors due to proportion estimations should not be neglected. They further confirm that the accuracy of response designs depends on the class proportions within the sampling units, with complex landscapes being more prone to errors. As a result, response designs where the number of sub-samples is predefined and fixed are inefficient. To guarantee high accuracy standards of validation data with minimum data collection effort, we propose a new method to adapt the number of sub-samples for each sample during the validation process. In practice, sub-samples are incrementally selected and labeled until the estimated class proportions reach the desired level of confidence. As a result, less effort is spent on labeling univocal cases and the spared effort can be allocated to more ambiguous cases. This increases the reliability of reference data and of subsequent accuracy assessment. Across our study site, we demonstrated that such an approach could reduce the labeling effort by 50% to 75%, with greater gains in homogeneous landscapes. We contend that adopting this optimization approach will not only increase the efficiency of reference data collection, but will also help deliver more reliable accuracy estimates to the user community.

High temporal resolution of leaf area data improves empirical estimation of grain yield

Empirical yield estimation from satellite data has long lacked suitable combinations of spatial and temporal resolutions. Consequently, the selection of metrics, i.e., temporal descriptors that predict grain yield, has likely been driven by practicality and data availability rather than by systematic targetting of critically sensitive periods as suggested by knowledge of crop physiology. The current trend towards hyper-temporal data raises two questions: How does temporality affect the accuracy of empirical models? Which metrics achieve optimal performance? We followed an in silico approach based on crop modelling which can generate any observation frequency, explore a range of growing conditions and reduce the cost of measuring yields in situ. We simulated wheat crops across Australia and regressed six types of metrics derived from the resulting time series of Leaf Area Index (LAI) against wheat yields. Empirical models using advanced LAI metrics achieved national relevance and, contrary to simple metrics, did not benefit from the addition of weather information. This suggests that they already integrate most climatic effects on yield. Simple metrics remained the best choice when LAI data are sparse. As we progress into a data-rich era, our results support a shift towards metrics that truly harness the temporal dimension of LAI data.

Mapping global development potential for renewable energy, fossil fuels, mining and agriculture sectors

Mapping suitable land for development is essential to land use planning efforts that aim to model, anticipate, and manage trade-offs between economic development and the environment. Previous land suitability assessments have generally focused on a few development sectors or lack consistent methodologies, thereby limiting our ability to plan for cumulative development pressures across geographic regions. Here, we generated 1-km spatially-explicit global land suitability maps, referred to as "development potential indices" (DPIs), for 13 sectors related to renewable energy (concentrated solar power, photovoltaic solar, wind, hydropower), fossil fuels (coal, conventional and unconventional oil and gas), mining (metallic, non-metallic), and agriculture (crop, biofuels expansion). To do so, we applied spatial multi-criteria decision analysis techniques that accounted for both resource potential and development feasibility. For each DPI, we examined both uncertainty and sensitivity, and spatially validated the map using locations of planned development. We illustrate how these DPIs can be used to elucidate potential individual sector expansion and cumulative development patterns.

Managing the middle: A shift in conservation priorities based on the global human modification gradient

An increasing number of international initiatives aim to reconcile development with conservation. Crucial to successful implementation of these initiatives is a comprehensive understanding of the current ecological condition of landscapes and their spatial distributions. Here, we provide a cumulative measure of human modification of terrestrial lands based on modeling the physical extents of 13 anthropogenic stressors and their estimated impacts using spatially explicit global datasets with a median year of 2016. We quantified the degree of land modification and the amount and spatial configuration of low modified lands (i.e., natural areas relatively free from human alteration) across all ecoregions and biomes. We identified that fewer unmodified lands remain than previously reported and that most of the world is in a state of intermediate modification, with 52% of ecoregions classified as moderately modified. Given that these moderately modified ecoregions fall within critical land use thresholds, we propose that they warrant elevated attention and require proactive spatial planning to maintain biodiversity and ecosystem function before important environmental values are lost.

Estimating the global distribution of field size using crowdsourcing

There is an increasing evidence that smallholder farms contribute substantially to food production globally, yet spatially explicit data on agricultural field sizes are currently lacking. Automated field size delineation using remote sensing or the estimation of average farm size at subnational level using census data are two approaches that have been used. However, both have limitations, for example, automatic field size delineation using remote sensing has not yet been implemented at a global scale while the spatial resolution is very coarse when using census data. This paper demonstrates a unique approach to quantifying and mapping agricultural field size globally using crowdsourcing. A campaign was run in June 2017, where participants were asked to visually interpret very high resolution satellite imagery from Google Maps and Bing using the Geo-Wiki application. During the campaign, participants collected field size data for 130 K unique locations around the globe. Using this sample, we have produced the most accurate global field size map to date and estimated the percentage of different field sizes, ranging from very small to very large, in agricultural areas at global, continental, and national levels. The results show that smallholder farms occupy up to 40% of agricultural areas globally, which means that, potentially, there are many more smallholder farms in comparison with the two different current global estimates of 12% and 24%. The global field size map and the crowdsourced data set are openly available and can be used for integrated assessment modeling, comparative studies of agricultural dynamics across different contexts, for training and validation of remote sensing field size delineation, and potential contributions to the Sustainable Development Goal of Ending hunger, achieve food security and improved nutrition and promote sustainable agriculture.

Characterizing the Spatial and Temporal Availability of Very High Resolution Satellite Imagery in Google Earth and Microsoft Bing Maps as a Source of Reference Data

Very high resolution (VHR) satellite imagery from Google Earth and Microsoft Bing Maps is increasingly being used in a variety of applications from computer sciences to arts and humanities. In the field of remote sensing, one use of this imagery is to create reference data sets through visual interpretation, e.g., to complement existing training data or to aid in the validation of land-cover products. Through new applications such as Collect Earth, this imagery is also being used for monitoring purposes in the form of statistical surveys obtained through visual interpretation. However, little is known about where VHR satellite imagery exists globally or the dates of the imagery. Here we present a global overview of the spatial and temporal distribution of VHR satellite imagery in Google Earth and Microsoft Bing Maps. The results show an uneven availability globally, with biases in certain areas such as the USA, Europe and India, and with clear discontinuities at political borders. We also show that the availability of VHR imagery is currently not adequate for monitoring protected areas and deforestation, but is better suited for monitoring changes in cropland or urban areas using visual interpretation.

A large-area, spatially continuous assessment of land cover map error and its impact on downstream analyses

Land cover maps increasingly underlie research into socioeconomic and environmental patterns and processes, including global change. It is known that map errors impact our understanding of these phenomena, but quantifying these impacts is difficult because many areas lack adequate reference data. We used a highly accurate, high-resolution map of South African cropland to assess (1) the magnitude of error in several current generation land cover maps, and (2) how these errors propagate in downstream studies. We first quantified pixel-wise errors in the cropland classes of four widely used land cover maps at resolutions ranging from 1 to 100 km, and then calculated errors in several representative "downstream" (map-based) analyses, including assessments of vegetative carbon stocks, evapotranspiration, crop production, and household food security. We also evaluated maps' spatial accuracy based on how precisely they could be used to locate specific landscape features. We found that cropland maps can have substantial biases and poor accuracy at all resolutions (e.g., at 1 km resolution, up to ∼45% underestimates of cropland (bias) and nearly 50% mean absolute error (MAE, describing accuracy); at 100 km, up to 15% underestimates and nearly 20% MAE). National-scale maps derived from higher-resolution imagery were most accurate, followed by multi-map fusion products. Constraining mapped values to match survey statistics may be effective at minimizing bias (provided the statistics are accurate). Errors in downstream analyses could be substantially amplified or muted, depending on the values ascribed to cropland-adjacent covers (e.g., with forest as adjacent cover, carbon map error was 200%-500% greater than in input cropland maps, but ∼40% less for sparse cover types). The average locational error was 6 km (600%). These findings provide deeper insight into the causes and potential consequences of land cover map error, and suggest several recommendations for land cover map users.

A global reference database of crowdsourced cropland data collected using the Geo-Wiki platform

A global reference data set on cropland was collected through a crowdsourcing campaign using the Geo-Wiki crowdsourcing tool. The campaign lasted three weeks, with over 80 participants from around the world reviewing almost 36,000 sample units, focussing on cropland identification. For quality assessment purposes, two additional data sets are provided. The first is a control set of 1,793 sample locations validated by students trained in satellite image interpretation. This data set was used to assess the quality of the crowd as the campaign progressed. The second data set contains 60 expert validations for additional evaluation of the quality of the contributions. All data sets are split into two parts: the first part shows all areas classified as cropland and the second part shows cropland average per location and user. After further processing, the data presented here might be suitable to validate and compare medium and high resolution cropland maps generated using remote sensing. These could also be used to train classification algorithms for developing new maps of land cover and cropland extent.

Validation of Automatically Generated Global and Regional Cropland Data Sets: The Case of Tanzania

There is a need to validate existing global cropland maps since they are used for different purposes including agricultural monitoring and assessment. In this paper we validate three recent global products (ESA-CCI, GlobeLand30, FROM-GC) and one regional product (Tanzania Land Cover 2010 Scheme II) using a validation data set that was collected by students through the Geo-Wiki tool. The ultimate aim was to understand the usefulness of these products for agricultural monitoring. Data were collected wall-to-wall for Kilosa district and for a sample across Tanzania. The results show that the amount of and spatial extent of cropland in the different products differs considerably from 8% to 42% for Tanzania, with similar values for Kilosa district. The agreement of the validation data with the four different products varied between 36% and 54% and highlighted that cropland is overestimated by the ESA-CCI and underestimated by FROM-GC. The validation data were also analyzed for consistency between the student interpreters and also compared with a sample interpreted by five experts for quality assurance. Regarding consistency between the students, there was more than 80% agreement if one difference in cropland category was considered (e.g., between low and medium cropland) while most of the confusion with the experts was also within one category difference. In addition to the validation of current cropland products, the data set collected by the students also has potential value as a training set for improving future cropland products.