Statistical tests for non-independent partitions of large autocorrelated datasets

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

doi:10.1016/j.mex.2022.101660

Cited by 3 publications

(7 citation statements)

References 7 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We defined the magnitude of the spatial correlation in each trend dataset as the unexplained variation in trends among pixels where the magnitude of autocorrelation depends on the distance between pixels (i.e., pixels located closer are more similar):

cor (}{,, e_{i}, e_{j}) goodbreak= nugget goodbreak+ ()(, 1 goodbreak- nugget) \exp^{- {(\frac{d_{italicij}}{range})}^{shape}},

where cor{ e i , e j } is the correlation between the random errors in pixels i and j , e i and e j , d ij is the distance between pixels, and nugget, range, and shape are parameters that determine the dependence of the correlation on distance. We calculated the range and shape of spatial correlation for each trend map, which give the distance beyond which two pixels are uncorrelated and the shape of the decay function, respectively (Ives et al, 2021, 2022).…”

Section: Methodsmentioning

confidence: 99%

“…Developed by Ives et al (2021Ives et al ( , 2022 PARTS is a three-step statistical approach that allows for the analysis of trends in time series of raster data while accounting for temporal and spatial autocorrelation and in combination with additional covariates (e.g., meteorological data, or land cover information). First, by using an autoregressive model, PARTS derives temporally uncorrelated trends for each pixel in the time series.…”

Section: Parts: Accounting For Temporal and Spatial Autocorrelation I...mentioning

confidence: 99%

See 1 more Smart Citation

Beyond “greening” and “browning”: Trends in grassland ground cover fractions across Eurasia that account for spatial and temporal autocorrelation

Lewińska

Ives

Morrow

et al. 2023

Global Change Biology

Self Cite

View full text Add to dashboard Cite

Grassland ecosystems cover up to 40% of the global land area and provide many ecosystem services directly supporting the livelihoods of over 1 billion people. Monitoring long-term changes in grasslands is crucial for food security, biodiversity conservation, achieving Land Degradation Neutrality goals, and modeling the global carbon budget.Although long-term grassland monitoring using remote sensing is extensive, it is typically based on a single vegetation index and does not account for temporal and spatial autocorrelation, which means that some trends are falsely identified while others are missed. Our goal was to analyze trends in grasslands in Eurasia, the largest continuous grassland ecosystems on Earth. To do so, we calculated Cumulative Endmember Fractions (annual sums of monthly ground cover fractions) derived from MODIS 2002-2020 time series, and applied a new statistical approach PARTS that explicitly accounts for temporal and spatial autocorrelation in trends. We examined trends in green vegetation, non-photosynthetic vegetation, and soil ground cover fractions considering their independent change trajectories and relations among fractions over time. We derived temporally uncorrelated pixel-based trend maps and statistically tested whether observed trends could be explained by elevation, land cover, SPEI3, climate, country, and their combinations, all while accounting for spatial autocorrelation. We found no statistical evidence for a decrease in vegetation cover in grasslands in Eurasia. Instead, there was a significant map-level increase in non-photosynthetic vegetation across the region and local increases in green vegetation with a concomitant decrease in soil fraction. Independent environmental variables affected trends significantly, but effects varied by region. Overall, our analyses show in a statistically robust manner that Eurasian grasslands have changed considerably over the past two decades. Our approach enhances remote sensing-based monitoring of trends in grasslands so that underlying processes can be discerned.

show abstract

cor (}{,, e_{i}, e_{j}) goodbreak= nugget goodbreak+ ()(, 1 goodbreak- nugget) \exp^{- {(\frac{d_{italicij}}{range})}^{shape}},

Section: Methodsmentioning

confidence: 99%

Section: Parts: Accounting For Temporal and Spatial Autocorrelation I...mentioning

confidence: 99%

Beyond “greening” and “browning”: Trends in grassland ground cover fractions across Eurasia that account for spatial and temporal autocorrelation

Lewińska

Ives

Morrow

et al. 2023

Global Change Biology

Self Cite

View full text Add to dashboard Cite

show abstract

“…The remotePARTS package then calculated cross‐partition statistics from pairs of partitions and summarized the results with statistical tests that account for correlations among partitions. This method has been demonstrated to have good statistical performance and is suitable for hypothesis testing with large spatial or temporal datasets (Ives et al, 2022).…”

Section: Methodsmentioning

confidence: 99%

“…There is a large computational burden to fit models with a regular GLS given the large number of grid cells in this study (n = 12,700; which means a distance and covariance matrix with >161 million elements). To overcome these computational limitations, we conducted all spatial analyses with the R package "re-motePARTS" (Ives et al, 2021), which allows for the partitioning of large spatial datasets and a later integration of regression results derived from these data (Ives et al, 2022). Specifically, remotePARTS uses the "split and conquer" strategy to analyse random partitions of the data separately, then combines the statistical results from all partitions into an overall statistical test.…”

Section: We Examined Correlations Among Predictor Variables Withmentioning

confidence: 99%

See 1 more Smart Citation

Global patterns and drivers of raptor phylogenetic and functional diversity

Montaño‐Centellas

Baiser

McGrew

et al. 2022

Global Ecol Biogeogr

View full text Add to dashboard Cite

Aim Raptors, a highly threatened but ecologically important group of birds, have been recognized as a good proxy for overall biodiversity in conservation planning. However, previous work on raptor diversity focused predominantly on taxonomic diversity. Here, we assess the global patterns of raptor taxonomic, phylogenetic and functional diversity and their association with current and historical environmental factors. Location Global. Time period Present day. Major taxa studied Raptors. Methods We compiled information from distribution maps, global trait datasets and avian phylogenies for all extant raptors. We used generalized least squares (GLS) to assess the relationship between historical and contemporary environmental predictors and species richness, phylogenetic diversity [Faith's phylogenetic diversity index (Faith's PD), mean pairwise distance (MPD) and mean nearest taxon distance (MNTD)] and functional diversity of traits related to raptor morphology, lifestyles, diet, foraging and vagility, while controlling for spatial autocorrelation. Results Raptor taxonomic diversity peaked in tropical regions and nearby mountain ranges. After controlling for species richness, species‐poor assemblages in high latitudes and deserts were more phylogenetically and functionally diverse than expected by chance. In species‐rich assemblages, diet and foraging traits had greater variation, whereas morphological traits had less variation than expected, suggesting that species packing promoted adaptive radiation in these assemblages. Historical climate influenced phylogenetic diversity and functional diversity of morphological, foraging and diet traits, leaving a signal of evolutionary history on modern assemblages. Human footprint was also an important driver of MPD and of niche and vagility trait functional diversity, with higher phylogenetic diversity in disturbed areas and with higher functional diversity in regions with intermediate levels of disturbance. Main conclusions Both palaeoclimate and contemporary environmental conditions are important drivers of raptor phylogenetic and functional diversity. We found large mismatches among taxonomic, functional and phylogenetic diversity, demonstrating how different processes filter lineages and species traits shaping raptor assemblages. Our results highlight the need to consider multiple dimensions of diversity to inform conservation planning better when using raptors as umbrella species.

show abstract

The phenology of the spring phytoplankton bloom in the North Atlantic does not trend with temperature

Friedland,

Nielsen,

Record

et al. 2024

Elem Sci Anth

View full text Add to dashboard Cite

Climate change is anticipated to alter the phenology of phytoplankton blooms in the ocean, making their recent dynamics of interest to inform models of future ocean states. We characterized temperature change in the North Atlantic using metrics that track the patterns of sea surface water temperature (SST) defined by quantiles. To complement these thermal indicators, we estimated a thermal phenology index in the form of the date of the spring transition, taken as the date that temperature achieved the long-term mean at a specific location. We then used ocean color data (1998–2022) and characterized spring bloom phenology using change point methods to derive bloom initiation, duration, magnitude, and intensity. The North Atlantic has warmed over recent decades, averaging a rate of increase of 0.27°C decade−1, yet throughout most of the basin, spring transition timing has remained constant, with the exception of small areas with either delayed or advanced transitions. There were no clear trends in bloom start or duration in the North Atlantic, indicating that spring bloom phenology was independent of climate-driven temperature change. Bloom magnitude and intensity trended downward in some North Atlantic continental shelf seas, indicating that increased temperatures may have had negative effects on overall bloom productivity. However, exclusive of the areas where the bloom parameters were trending, there was a decrease in magnitude and intensity with warmer winter temperatures, suggesting that the inter-annual variability of these parameters may be affected by thermal conditions at the onset of the bloom. While temperature has increased in the North Atlantic, vernal light availability has remained unchanged, which may explain why spring bloom phenology has remained resistant to climate change. Consequently, it seems plausible that future climate change may have limited effects on spring bloom phenology, but could have substantial effects on overall phytoplankton production.

show abstract

Statistical tests for non-independent partitions of large autocorrelated datasets

Cited by 3 publications

References 7 publications

Beyond “greening” and “browning”: Trends in grassland ground cover fractions across Eurasia that account for spatial and temporal autocorrelation

Beyond “greening” and “browning”: Trends in grassland ground cover fractions across Eurasia that account for spatial and temporal autocorrelation

Global patterns and drivers of raptor phylogenetic and functional diversity

The phenology of the spring phytoplankton bloom in the North Atlantic does not trend with temperature

Contact Info

Product

Resources

About