Optimum machine learning algorithm selection for forecasting vegetation indices: MODIS NDVI &amp; EVI

Roy, Bishal

doi:10.1016/j.rsase.2021.100582

Cited by 17 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Vegetation indices (NDVI and EVI) extracted from the 2001 to 2018 MODIS dataset have also been used to forecast their values in 2019 using Vector Regression, Random Forest (RF), and Linear and Polynomial Regression (Roy, 2021). For predicting maize yield from land surface temperature (LST) and NDVI in Pakistan, Ahmad et al (2020a) applied the K-Nearest Neighbor clustering machine learning model.…”

Section: Use Of Machine Learning For Spatiotemporal Data Fusion Ndvi-...mentioning

confidence: 99%

“…For instance, CNN and RF display good performance in vegetation growth predictions from NDVI (Ayhan et al, 2020;Li et al, 2021;Mishra and Shahi, 2021;Ferchichi et al, 2022). The performance of machine learning models can be evaluated through a range of approaches, including Root Mean Square Error (RMSE), coefficient of determinates (R 2) , Pearson correlation (R), and structural similarity (SSIM), which have been used by Rhif et al (2020), Ahmad et al (2020b), Arab et al (2021), Htitiou et al (2021), Mishra andShahi (2021), andRoy (2021). Htitiou et al (2021) use NDVI values extracted from spatial transects created across the study site to compare the performance of Very Deep Super-Resolution (VDSR) against the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) and the flexible spatiotemporal data fusion (FSDAF) method in producing high resolution NDVI time series datasets.…”

Section: Use Of Machine Learning For Spatiotemporal Data Fusion Ndvi-...mentioning

confidence: 99%

See 1 more Smart Citation

Can remote sensing enable a Biomass Climate Adaptation Index for agricultural systems?

et al. 2022

View full text Add to dashboard Cite

Systematic tools and approaches for measuring climate change adaptation at multiple scales of spatial resolution are lacking, limiting measurement of progress toward the adaptation goals of the Paris Agreement. In particular, there is a lack of adaptation measurement or tracking systems that are coherent (measuring adaptation itself), comparable (allowing comparisons across geographies and systems), and comprehensive (are supported by the necessary data). In addition, most adaptation measurement efforts lack an appropriate counterfactual baseline to assess the effectiveness of adaptation-related interventions. To address this, we are developing a “Biomass Climate Adaptation Index” (Biomass CAI) for agricultural systems, where climate adaptation progress across multiple scales can be measured by satellite remote sensing. The Biomass CAI can be used at global, national, landscape and farm-level to remotely monitor agri-biomass productivity associated with adaptation interventions, and to facilitate more tailored “precision adaptation”. The Biomass CAI places focus on decision-support for end-users to ensure that the most effective climate change adaptation investments and interventions can be made in agricultural and food systems.

show abstract

Section: Use Of Machine Learning For Spatiotemporal Data Fusion Ndvi-...mentioning

confidence: 99%

Section: Use Of Machine Learning For Spatiotemporal Data Fusion Ndvi-...mentioning

confidence: 99%

Can remote sensing enable a Biomass Climate Adaptation Index for agricultural systems?

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Using remotely sensed satellite imagery, a change in surface temperature may be observed, which is an indicator of impervious surface build-up [ 26 ]. The distribution of LST is greatly influenced by the presence of natural vegetation [ 15 , 27 , 28 , 29 ]. To study variations in LST, the Normalized Difference Vegetation Index (NDVI) is often used [ 12 , 30 , 31 ].…”

Section: Introductionmentioning

confidence: 99%

Examining the relationship between land surface temperature and landscape features using spectral indices with Google Earth Engine

Roy

Bari²

2022

Heliyon

Self Cite

View full text Add to dashboard Cite

“…Machine learning (ML) methods have been used for downscaling and gap-filling purposes in remote sensing products and can be seen as one tool that may lead to the production of high quality remote sensing products (Zhu et al, 2022;Zeng et al, 2013). Furthermore, ML methods have been successfully applied to a wide range of drought-related (Hauswirth et al, 2021;Shamshirband et al, 2020;Tufaner and Özbeyaz, 2020;Shen et al, 2019;Das et al, 2020;Hauswirth et al, 2022) and remotely sensed vegetation studies (Roy, 2021;Li et al, 2021b;Reichstein et al, 2019). Compared to conventional statistical downscaling techniques, ML is considered the superior alternative; given that no strict statistical assumptions are required, complex and non-linear relationships are well captured and provide high precision (Ebrahimy et al, 2021).…”

mentioning

confidence: 99%

“…Gap-filling can also be achieved by relying on the RF to predict values where data is sparse or missing (Wang et al, 2022). These studies have highlighted that ML methods can accurately predict the dynamics of vegetation (Roy, 2021;Gensheimer et al, 2022). However, studies applying ML methods to global vegetation dynamics concerning drought conditions are less prominent (Li et al, 2021b;Zhang et al, 2021b;Chen et al, 2021).…”

mentioning

confidence: 99%

Machine learning and Global Vegetation: Random Forests for Downscaling and Gapfilling

Jaarsveld

Hauswirth²,

Wanders³

2023

Preprint

View full text Add to dashboard Cite

Abstract. Drought is a devastating natural disaster, where water shortage often manifests itself in the health of vegetation. Unfortunately, it is difficult to obtain high-resolution vegetation drought impact, which is spatially and temporally consistent. While remotely sensed products can provide part of this information, they often suffer from data gaps and limitations in spatial or temporal resolutions. A persistent feature among remote sensing products is tradeoffs between spatial resolution and revisiting times, where high temporal resolution is met by coarse spatial resolution and vice verse. Machine learning methods have been successfully applied in a wide range of remote sensing and hydrological studies. However, global applications to resolve drought impacts on vegetation dynamics still need to be made available, while there is significant potential for such a product to aid improved drought impact monitoring. To this end, this study predicted global vegetation dynamics based on the Enhanced Vegetation Index (evi) and the popular Random Forest algorithm (RF) at 0.1°. We assessed the applicability of RF as a gap filling and downscaling tool to generate spatial and temporal consistent global evi estimates. To do this, we trained an RF regressor with 0.1° evi data using a host of features indicative of water and energy balances experienced by vegetation and we evaluated the performance of this new product. Next, to test whether the RF is robust in terms of spatial resolution, we downscale global evi, the model trained on 0.1° data is used to predict evi at 0.01° resolution. The results show that the RF can capture global evi dynamics at both the 0.1° (RMSE: 0.02–0.4) and at the finer 0.01° (RMSE: 0.04–0.6) resolution. Overall errors were higher in the down-scaled 0.01° compared to the 0.1° product. Yet, relative increases remained small, thus demonstrating that RF can be used to create downscaled and temporally consistent evi products. Additional error analysis reveals that errors vary spatiotemporally, with underrepresented landcover types and periods of extreme vegetation conditions having the highest errors. Finally, this model is used to produce global spatially continuous evi products at both the 0.1° and 0.01° spatial resolution for 2003–2013 at an 8-day frequency.

show abstract

Optimum machine learning algorithm selection for forecasting vegetation indices: MODIS NDVI & EVI

Cited by 17 publications

References 17 publications

Can remote sensing enable a Biomass Climate Adaptation Index for agricultural systems?

Can remote sensing enable a Biomass Climate Adaptation Index for agricultural systems?

Examining the relationship between land surface temperature and landscape features using spectral indices with Google Earth Engine

Machine learning and Global Vegetation: Random Forests for Downscaling and Gapfilling

Contact Info

Product

Resources

About