Statistical inference for trends in spatiotemporal data

Ives, Anthony R.; Zhu, Likai; Wang, Fangfang; Zhu, Jun; Morrow, Clay J.; Radeloff, Volker C.

doi:10.1016/j.rse.2021.112678

Cited by 30 publications

(38 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, statistical models, such as the one used in our study, applied to calculate trends, result in uncertainties, namely, Type I (false positive) and Type II errors (false negative). Type I errors can be easily propagated due to temporal autocorrelation, thus giving rise to trend patterns that might not depict any real change [49]. Unless there is an independent reference data set available, which is rarely the case with satellite time series, direct validation, and accuracy assessment of results is not possible.…”

Section: Discussionmentioning

confidence: 99%

Place-Based Analysis of Satellite Time Series Shows Opposing Land Change Patterns in the Copperbelt Region of Zambia

Munawar

Röder

Syampungani

et al. 2022

Forests

View full text Add to dashboard Cite

The process of land degradation needs to be understood at various spatial and temporal scales in order to protect ecosystem services and communities directly dependent on it. This is especially true for regions in sub-Saharan Africa, where socio economic and political factors exacerbate ecological degradation. This study identifies spatially explicit land change dynamics in the Copperbelt province of Zambia in a local context using satellite vegetation index time series derived from the MODIS sensor. Three sets of parameters, namely, monthly series, annual peaking magnitude, and annual mean growing season were developed for the period 2000 to 2019. Trend was estimated by applying harmonic regression on monthly series and linear least square regression on annually aggregated series. Estimated spatial trends were further used as a basis to map endemic land change processes. Our observations were as follows: (a) 15% of the study area dominant in the east showed positive trends, (b) 3% of the study area dominant in the west showed negative trends, (c) natural regeneration in mosaic landscapes (post shifting cultivation) and land management in forest reserves were chiefly responsible for positive trends, and (d) degradation over intact miombo woodland and cultivation areas contributed to negative trends. Additionally, lower productivity over areas with semi-permanent agriculture and shift of new encroachment into woodlands from east to west of Copperbelt was observed. Pivot agriculture was not a main driver in land change. Although overall greening trends prevailed across the study site, the risk of intact woodlands being exposed to various disturbances remains high. The outcome of this study can provide insights about natural and assisted landscape restoration specifically addressing the miombo ecoregion.

show abstract

Section: Discussionmentioning

confidence: 99%

Place-Based Analysis of Satellite Time Series Shows Opposing Land Change Patterns in the Copperbelt Region of Zambia

Munawar

Röder

Syampungani

et al. 2022

Forests

View full text Add to dashboard Cite

show abstract

“…The spatiotemporal analyses follow the approach presented in Ives et al. [9] for analyzing time trends in remote-sensing data. The approach involves first fitting a time-series model to the time series in each pixel on a map and obtaining the estimate of the time trend.…”

Section: Methods Detailsmentioning

confidence: 99%

“…The former is useful, because it allows analytical solutions, whereas the second has better statistical properties and is therefore preferentially used for the analyses in Ives et al. [9] .…”

Section: Methods Detailsmentioning

confidence: 99%

“…For the example dataset from Alaska analyzed in Ives et al. [9] , the time series have length T = 34, and estimates of β l had a mean of 0.40 and a standard deviation of 0.23. The simulations in Table 2 for AR(1) regression suggest that using cor[ε l ( t ), ε k ( t )] to approximate cor[

l ,

k ] will lead to a small overestimate of cor[

l ,

k ], although this will not change the conclusions; this is discussed in detail in Ives et al.…”

Section: Methods Detailsmentioning

confidence: 99%

See 1 more Smart Citation

Statistical tests for non-independent partitions of large autocorrelated datasets

et al. 2022

Self Cite

View full text Add to dashboard Cite

“…In general, data with a finer spatial resolution possess more information about the seasonality and phenology properties of vegetation [38][39][40]. Meanwhile, data with a fine spatial resolution have the problem of providing a larger amount of data, having a slow computation speed, and being time-consuming [41]. The coarse spatial resolution data is more suitable for monitoring phenology at a landscape scale [42].…”

Section: Introductionmentioning

confidence: 99%

Exploring the Applicability and Scaling Effects of Satellite-Observed Spring and Autumn Phenology in Complex Terrain Regions Using Four Different Spatial Resolution Products

et al. 2021

View full text Add to dashboard Cite

The information on land surface phenology (LSP) was extracted from remote sensing data in many studies. However, few studies have evaluated the impacts of satellite products with different spatial resolutions on LSP extraction over regions with a heterogeneous topography. To bridge this knowledge gap, this study took the Loess Plateau as an example region and employed four types of satellite data with different spatial resolutions (250, 500, and 1000 m MODIS NDVI during the period 2001–2020 and ~10 km GIMMS3g during the period 1982–2015) to investigate the LSP changes that took place. We used the correlation coefficient (r) and root mean square error (RMSE) to evaluate the performances of various satellite products and further analyzed the applicability of the four satellite products. Our results showed that the MODIS-based start of the growing season (SOS) and end of the growing season (EOS) were highly correlated with the ground-observed data with r values of 0.82 and 0.79, respectively (p < 0.01), while the GIMMS3g-based phenology signal performed badly (r < 0.50 and p > 0.05). Spatially, the LSP that was derived from the MODIS products produced more reasonable spatial distributions. The inter-annual averaged MODIS SOS and EOS presented overall advanced and delayed trends during the period 2001–2020, respectively. More than two-thirds of the SOS advances and EOS delays occurred in grasslands, which determined the overall phenological changes across the entire Loess Plateau. However, both inter-annual trends of SOS and EOS derived from the GIMMS3g data were opposite to those seen in the MODIS results. There were no significant differences among the three MODIS datasets (250, 500, and 1000 m) with regard to a bias lower than 2 days, RMSE lower than 1 day, and correlation coefficient greater than 0.95 (p < 0.01). Furthermore, it was found that the phenology that was derived from the data with a 1000 m spatial resolution in the heterogeneous topography regions was feasible. Yet, in forest ecosystems and areas with an accumulated temperature ≥10 °C, the differences in phenological phase between the MODIS products could be amplified.

show abstract

Statistical inference for trends in spatiotemporal data

Cited by 30 publications

References 64 publications

Place-Based Analysis of Satellite Time Series Shows Opposing Land Change Patterns in the Copperbelt Region of Zambia

Place-Based Analysis of Satellite Time Series Shows Opposing Land Change Patterns in the Copperbelt Region of Zambia

Statistical tests for non-independent partitions of large autocorrelated datasets

Exploring the Applicability and Scaling Effects of Satellite-Observed Spring and Autumn Phenology in Complex Terrain Regions Using Four Different Spatial Resolution Products

Contact Info

Product

Resources

About