The dataset consists of composited seasonal surface reflectance images (4 seasons per year) created from the full time series of Sentinel-2 imagery. The imagery has been composited over a season to produce imagery which is representative of that period, using techniques which will reduce contamination by cloud and other problems. This creates a regular time series of reflectance values which captures the variability at seasonal time scales. The benefits are a regular time series with minimal missing data or contamination from various sources of noise as well as data reduction. Each season has exactly one value (per band) for each pixel (or is null, i.e., missing), and the value for that season is assumed to be the representative of the whole season. The algorithm is based on the medoid (in reflectance space) over the time period (the medoid is a multi-dimensional analogue of the median), which is robust against extreme values. The seasonal surface reflectance is of the 6 TM-like bands (Blue, Green, Red, NIR, SWIR1, SWIR2), all at 10 m resolution. This dataset is intended to be a 10 m equivalent of the Landsat surface reflectance, using only Sentinel-2. The two 20m bands are resampled using cubic convolution. The pixel values are scaled reflectance, as 16-bit integers. To retrieve physical reflectance values, the pixel values should be multiplied by 0.0001.

The dataset consists of composited seasonal surface reflectance images (4 seasons per year) created from the full time series of Landsat TM/ETM+/OLI imagery. The imagery has been composited over a season to produce imagery which is representative of that period, using techniques which will reduce contamination by cloud and other problems. This creates a regular time series of reflectance values which captures the variability at seasonal time scales. The benefits are a regular time series with minimal missing data or contamination from various sources of noise as well as data reduction. Each season has exactly one value (per band) for each pixel (or is null, i.e., missing), and the value for that season is assumed to be the representative of the whole season. The algorithm is based on the medoid (in reflectance space) over the time period (the medoid is a multi-dimensional analogue of the median), which is robust against extreme values.

The seasonal dynamic reference cover method images are created using a modified version of the dynamic reference cover method developed by <a href="https://doi.org/10.1016/j.rse.2012.02.021">Bastin et al (2012) </a>. This approach calculates a minimum ground cover image over all years to identify locations of most persistent ground cover in years with the lowest rainfall, then uses a moving window approach to calculate the difference between the window's central pixel and its surrounding reference pixels. The output is a difference image between the cover amount of a pixel's reference pixels and the actual cover at that pixel for the season being analysed. Negative values indicate pixels which have less cover than the reference pixels. The main differences between this method and the original method are that this method uses seasonal fractional ground cover rather than the preceding ground cover index (GCI) and this method excludes cleared areas and certain landforms (undulating slopes), which are considered unsuitable for use as reference pixels. This product is based upon the JRSRP Fractional Cover 3.0 algorithm.

The seasonal fractional ground cover product is a spatially explicit raster product that shows the proportion of bare ground, green and non-green ground cover at medium resolution (30 m per-pixel) for each 3-month calendar season. It is derived directly from the seasonal fractional cover product, also produced by Queensland's Remote Sensing Centre. A 3 band (byte) image is produced: band 1 – bare ground fraction (in percent), band 2 - green vegetation fraction (in percent), band 3 – non-green vegetation fraction (in percent). The no data value is 255.</br> The seasonal fractional cover product predicts vegetation cover, but does not distinguish tree and mid-level woody foliage and branch cover from green and dry ground cover. As a result, in areas with even minimal tree cover (>15%), estimates of ground cover become uncertain.</br> With the development of the fractional cover time-series, it has become possible to derive an estimate of ‘persistent green’ based on time-series analysis. The persistent green vegetation product provides an estimate of the vertically-projected green-vegetation fraction where vegetation is deemed to persist over time. These areas are nominally woody vegetation. This separation of the 'persistent green' from the fractional cover product, allows for the adjustment of the underlying spectral signature of the fractional cover image and the creation of a resulting 'true' ground cover estimate for each season. The estimates of cover are restricted to areas of <60% woody vegetation.

The seasonal fractional cover product shows representative values for the proportion of bare, green and non-green cover, created from a time series of Sentinel-2 imagery. It is a spatially explicit raster product, which predicts vegetation cover at medium resolution (10 m per-pixel) for each 3-month calendar season. The green and non-green fractions may include a mix of woody and non-woody vegetation. This model was originally developed for Landsat imagery, but has been adapted for Sentinel-2 imagery to produce a 10 m resolution equivalent product. A 3 band (byte) image is produced: band 1 – bare ground fraction (in percent), band 2 - green vegetation fraction (in percent), band 3 – non-green vegetation fraction (in percent). The no data value is 255.

<p>This dataset shows the crops grown in Queensland's main cropping areas, for the winter and summer growing-seasons, from 1988 to the current year. The winter growing-season is defined as June to October, and the summer growing-season is November to May. The basis of the maps is imagery from the (when available) Landsat-5 TM, Landsat-7 ETM+, Landsat-(8,9) OLI, and Sentinel-2(A,B) satellites; MODIS MOD13Q1 imagery was used as a backup in the case of large, temporal data gaps. Clusters of temporally similar pixels, termed 'segments', were identified in the imagery for each growing season, and served as an approximation of field boundaries. Per-segment phenological information, derived from the satellite imagery, was then combined with a tiered, tree-based statistical classifier, using >10000 field observations as training data, and >4000 independent observations for validation. The dataset supersedes a former crop-mapping effort <a href ="https://doi.org/10.3390/rs8040312">(Schmidt et al., 2016)</a>.</p> <p>Each season has 2 maps: an end-of-season prediction and a mid-season prediction. The mid-season prediction is labelled "_vInterim" to indicate that it is based on a relatively short time series, and should be used with caution.</p> <p>For optimum display symbology files have been provided for both QGIS and ArcGIS.</p>

Two fractional cover decile products, green cover and total cover, are currently produced from the historical timeseries of seasonal fractional cover images. These products compare, at the per-pixel level, the level of cover for the specific season of interest against the long term cover for that same season. For each pixel, all cover values for the relevant seasons within a baseline period (1988-2013) are classified into deciles. The cover value for the pixel in the season of interest is then classified according to the decile in which it falls. This product is based upon the JRSRP Fractional Cover 3.0 algorithm.

The seasonal fractional cover product shows representative values for the proportion of bare, green and non-green cover across a season. It is a spatially explicit raster product, which predicts vegetation cover at medium resolution (30 m per-pixel) for each 3-month calendar season. The green and non-green fractions may include a mix of woody and non-woody vegetation. A 3 band (byte) image is produced: band 1 – bare ground fraction (in percent), band 2 - green vegetation fraction (in percent), band 3 – non-green vegetation fraction (in percent). The no data value is 255.

The Sentinel-2 seasonal fractional ground cover product shows the proportion of bare ground, green and non-green ground cover and is derived directly from the Sentinel-2 seasonal fractional cover product, also produced by Queensland's Remote Sensing Centre. The seasonal fractional cover product is a spatially explicit raster product, which predicts vegetation cover at medium resolution (10&nbsp;m per-pixel) for each 3-month calendar season. However, the seasonal fractional cover product does not distinguish tree and mid-level woody foliage and branch cover from green and dry ground cover. As a result, in areas with even minimal tree cover (>15%), estimates of ground cover become uncertain. With the development of the fractional cover time-series, it has become possible to derive an estimate of ‘persistent green’ based on time-series analysis. The persistent green vegetation product provides an estimate of the vertically-projected green-vegetation fraction where vegetation is deemed to persist over time. These areas are nominally woody vegetation. This separation of the 'persistent green' from the fractional cover product, allows for the adjustment of the underlying spectral signature of the fractional cover image and the creation of a resulting 'true' ground cover estimate for each season. The estimates of cover are restricted to areas of <60% woody vegetation. Currently, the persistent green product has only been produced at 30&nbsp;m pixel resolution based on Landsat imagery, resulting in this Sentinel-2 seasonal ground cover product having a medium 30&nbsp;m pixel resolution also. This is an experimental product which has not been fully validated. This product is similar to the <a href="https://portal.tern.org.au/metadata/23884 ">Seasonal ground cover - Landsat, JRSRP algorithm Version 3.0, Australia Coverage</a> which is based on a different satellite sensor.

The TERN citizen science fauna monitoring pilot project aims to collect long-term monitoring data using the TERN standard monitoring protocol for birds. In May 2021, TERN launched a citizen science project designed to monitor bird biodiversity at the Samford Ecological Research Facility (SERF) in peri-urban Brisbane, Queensland, Australia, in collaboration with Queensland University of Technology and the Samford Eco-Corridor community group. Bird monitoring is undertaken quarterly by citizen scientists using TERN standardised monitoring protocols which are based on the Birds Australia systematic bird survey ‘2&nbsp;ha 20&nbsp;minute search’ method.