DEA Fractional Cover Percentiles (Landsat) Version 4.0.0
Bibliographic record
Abstract
Fractional Cover Percentiles (Landsat) estimate the 10th, 50th, and 90th percentiles independently for the green vegetation, non-green vegetation, and bare soil fractions observed in each calendar year from 1987.<br>The spatial extent is all Australia and the spatial resolution is 30 m x 30 m.<br>Percentiles provide an indicator of where an observation sits, relative to the rest of the observations for the pixel. For example, the 90th percentile is the value below which 90% of the observations fall. The 10th, 50th, and 90th percentiles represent low, median and high values in a distribution that are robust against outliers. These values can be used separately or combined to understand the land cover dynamics. For example, the three percentiles for the green cover fraction can serve as proxies for the minimum, typical and maximum green cover for a given year. Difference between the 10th and 90th percentiles provides an estimate of the magnitude of change within a year. A large range of values may be observed in the agricultural land for all cover types while high green cover and a small difference between 10th and 90th percentiles are expected for forest cover.A representative view of the landscape in a year can be obtained by combining the 50th percentiles, or the median values, for the three cover types.<br><strong>The statistics are calculated using the following satellites for the following periods of time:</strong>- 1987-1998 : Landsat 5 only- 1999 : Landsat 5 and Landsat 7- 2000-2002 : Landsat 7 only- 2003 : Landsat 5 and Landsat 7- 2004-2010 : Landsat 5 only- 2011-2012 : Landsat 7 only- 2013-2021 : Landsat 8 only- 2022 onwards: Landsat 8 and Landsat 9<br><strong>The values for this product are as follows:</strong>For the fractional cover bands (PV, NPV, BS)0-100 = fractional cover values that range between 0 and 100%<br><strong>Quality Assurance:</strong>This layer provides a breakdown of each FCP pixel between:- sufficient observations- insufficient observations dry- insufficient observations wetFor insufficient observations, these are pixels that have been masked out of the percentiles results e.g. NODATA, and provides an explanation as to why they have been masked out.<br><strong>Each product’s datasets is:</strong>- divided into tiles of 3200 x 3200 pixels, with a pixel size of 30 m x 30 m- presented in EPSG:3577 coordinate reference system<br><strong>Fractional Cover Masking</strong>DEA Water Observations are used to identify clear pixels from DEA Fractional Cover to be included in percentile calculation. A Fractional Cover observation is included if:<br>- it has corresponding DEA Water Observation information. If an observation within DEA Fractional Cover has no corresponding Water Observation, it is discarded. This can happen for ARD scenes that have a geometric quality assessment of greater than one, which occurs when there is poor geometric quality.<br>- the DEA Water Observation has the following characteristics: -- it is contiguous (data for all bands is present and valid), -- it is not saturated, -- it is not cloud, -- it is not cloud shadow, -- it is not terrain shadow, -- it is not low solar angle, -- it can be high slope, -- it is not wet, -- there are at least 3 clear and dry observations for the time period.<br>- No land/sea masking is applied.<br>- Observation dates for given percentiles are not captured.<br>Link to data: https://data.dea.ga.gov.au/?prefix=derivative/ga_ls_fc_pc_cyear_3/4-0-0/Link to DEA Knowledge Hub: https://knowledge.dea.ga.gov.au/data/product/dea-fractional-cover-percentiles-landsat/?tab=description<br><br>
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.010 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".