Mapping smallholder and large-scale cropland dynamics with a flexible classification system and pixel-based composites in an emerging frontier of Mozambique

Bey, Adia; Jetimane, Julieta; Lisboa, Sá Nogueira; Ribeiro, Natasha; Sitoe, Almeida; Meyfroidt, Patrick

doi:10.1016/j.rse.2019.111611

Cited by 59 publications

(45 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A large proportion of remote-sensing projects make some use of TD, typically created either using geolocated in situ data [46,60], by visually interpreting high and/or very high spatial-resolution imagery [26,61,62], or by interpreting the images to be classified/modeled themselves, e.g., [55,56,63,64]. Of these collection methods, HR/VHR image interpretation is increasingly common [65], particularly with the rise in crowdsourcing initiatives [22,66].…”

Section: Current Trends In Training Data (Td) Collectionmentioning

confidence: 99%

“…Previous research shows that the impact of TD error can be substantial but varied, suggesting that a more comprehensive approach to this issue is warranted. Furthermore, while TD and map reference data are often collected using the same approaches [55][56][57] and often subject to the same errors, the existing procedures to minimize and account for map reference errors [34,38,46,47] are not necessarily relevant for quantifying the impacts of TD error. The problems associated with TD error can be summarized as follows:1.The "big data" era vastly increases the demand for TD.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

et al. 2020

View full text Add to dashboard Cite

Remote sensing, or Earth Observation (EO), is increasingly used to understand Earth system dynamics and create continuous and categorical maps of biophysical properties and land cover, especially based on recent advances in machine learning (ML). ML models typically require large, spatially explicit training datasets to make accurate predictions. Training data (TD) are typically generated by digitizing polygons on high spatial-resolution imagery, by collecting in situ data, or by using pre-existing datasets. TD are often assumed to accurately represent the truth, but in practice almost always have error, stemming from (1) sample design, and (2) sample collection errors. The latter is particularly relevant for image-interpreted TD, an increasingly commonly used method due to its practicality and the increasing training sample size requirements of modern ML algorithms. TD errors can cause substantial errors in the maps created using ML algorithms, which may impact map use and interpretation. Despite these potential errors and their real-world consequences for map-based decisions, TD error is often not accounted for or reported in EO research. Here we review the current practices for collecting and handling TD. We identify the sources of TD error, and illustrate their impacts using several case studies representing different EO applications (infrastructure mapping, global surface flux estimates, and agricultural monitoring), and provide guidelines for minimizing and accounting for TD errors. To harmonize terminology, we distinguish TD from three other classes of data that should be used to create and assess ML models: training reference data, used to assess the quality of TD during data generation; validation data, used to iteratively improve models; and map reference data, used only for final accuracy assessment. We focus primarily on TD, but our advice is generally applicable to all four classes, and we ground our review in established best practices for map accuracy assessment literature. EO researchers should start by determining the tolerable levels of map error and appropriate error metrics. Next, TD error should be minimized during sample design by choosing a representative spatio-temporal collection strategy, by using spatially and temporally relevant imagery and ancillary data sources during TD creation, and by selecting a set of legend definitions supported by the data. Furthermore, TD error can be minimized during the collection of individual samples by using consensus-based collection strategies, by directly comparing interpreted training observations against expert-generated training reference data to derive TD error metrics, and by providing image interpreters with thorough application-specific training. We strongly advise that TD error is incorporated in model outputs, either directly in bias and variance estimates or, at a minimum, by documenting the sources and implications of error. TD should be fully documented and made available via an open TD repository, allowing others to replicate and assess its use. To guide researchers in this process, we propose three tiers of TD error accounting standards. Finally, we advise researchers to clearly communicate the magnitude and impacts of TD error on map outputs, with specific consideration given to the likely map audience.

show abstract

Section: Current Trends In Training Data (Td) Collectionmentioning

confidence: 99%

mentioning

confidence: 99%

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Numerous studies have found pixel-based compositing beneficial to reduce redundant information in large datasets, aggregate data into more distinct features, better characterize landscapes with different seasonal phenologies, and capitalize on all available cloud-and haze-free pixels 95,96,97 . Thus, we applied a median pixel method to generate seasonal composites, drawing on the findings of a recent study on the efficacy of this method in the study area 98…”

Section: Remote Sensingmentioning

confidence: 99%

The expansion of large-scale tree plantations: Detection, pathways, and development trade-offs

Bey

Meyfroidt

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Tree plantations can contribute to achieving several Sustainable Development Goals (SDGs) by creating job opportunities, storing carbon, and providing wood products that relieve pressure on natural forests. However, their impacts on SDGs are largely contingent upon the land uses they replace. Here we present a framework for understanding the development trade-offs associated with tree plantations and we introduce a new remote sensing technique to detect their expansion and pre-conversion land use. Using this approach in northern Mozambique, we found that 70% of 2001-2017 tree plantation expansion occurred on cropland, potentially exacerbating poverty, food insecurity and rural unemployment. The remainder occurred on natural vegetation, adversely affecting climate change mitigation efforts and life on land. Forty-one percent of plantation expansion occurred on land not designated for this land use. As tree plantations become more prevalent globally, monitoring their trajectories and expansion is critical to maximizing their benefits for people and the planet.

show abstract

“…The number of papers about GEE [ 26 , 30 , 31 , 32 ] in recent years has increased exponentially in several fields, such as: (i) forest and vegetation [ 33 , 34 ]; (ii) land use and land cover [ 35 , 36 , 37 ]; (iii) hydrology [ 38 , 39 , 40 ]; (iv) ecosystems and sustainability [ 41 ]; (v) agriculture [ 42 , 43 , 44 , 45 ]; (vi) climate [ 46 ]; (vii) urban sprawl [ 47 ]; (viii) hazards [ 48 , 49 ]; (ix) cultural heritage [ 50 ]. The use of GEE has allowed multiple tools to be created and shared for free.…”

Section: Introductionmentioning

confidence: 99%

Google Earth Engine as Multi-Sensor Open-Source Tool for Supporting the Preservation of Archaeological Areas: The Case Study of Flood and Fire Mapping in Metaponto, Italy

Fattore

Abate

Faridani

et al. 2021

Sensors

View full text Add to dashboard Cite

In recent years, the impact of Climate change, anthropogenic and natural hazards (such as earthquakes, landslides, floods, tsunamis, fires) has dramatically increased and adversely affected modern and past human buildings including outstanding cultural properties and UNESCO heritage sites. Research about protection/monitoring of cultural heritage is crucial to preserve our cultural properties and (with them also) our history and identity. This paper is focused on the use of the open-source Google Earth Engine tool herein used to analyze flood and fire events which affected the area of Metaponto (southern Italy), near the homonymous Greek-Roman archaeological site. The use of the Google Earth Engine has allowed the supervised and unsupervised classification of areas affected by flooding (2013–2020) and fire (2017) in the past years, obtaining remarkable results and useful information for setting up strategies to mitigate damage and support the preservation of areas and landscape rich in cultural and natural heritage.

show abstract

Mapping smallholder and large-scale cropland dynamics with a flexible classification system and pixel-based composites in an emerging frontier of Mozambique

Cited by 59 publications

References 56 publications

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

The expansion of large-scale tree plantations: Detection, pathways, and development trade-offs

Google Earth Engine as Multi-Sensor Open-Source Tool for Supporting the Preservation of Archaeological Areas: The Case Study of Flood and Fire Mapping in Metaponto, Italy

Contact Info

Product

Resources

About