Spatial and temporal dynamics of grass cover in the Amazon

README: Grass Coverage Area 1
INSERT LINK TO THE DATA ON DRYAD HERE

The data in this folder consists of a temporal raster stack of grass coverage in Area 1 of Tanguro Field Station in Mato Grosso, Brazil. Grass was documented in 2012, 2015, 2022, and 2025. Rasters were created from gridded observations of grass coverage in the plot and values were assigned according to species. It also contains a .csv with the % coverage of each species, for each year, for the Burn plots only (B1yr, B3yr). The shapefile for the plots is uploaded as well. Original data was collected by Divino Silverio and the team at the Tanguro Field Station.

Description of the data and file structure
There are four rasters/bands in the raster stack Grass_20122025.tif. Each band is one year (2012, 2015, 2022, 2025). The values in the raster are as follows:

0 = Forest
1 = Andropogon gayanus
2 = Aristida longifolia
3 = Grass-mix
4 = Pennisetum sp.
5 = Other non-grass
6 = Shaded Aristida longifolia*

Shaded Aristida longifolia is A.longifolia growing under forest canopy. It is a new classification added in 2025. Previous surveys did not quantify grass under tree cover.

For the shapefile, the treatment plots are B1Yr, B3Yr, and Control. B1Yr was burned once every year from 2004-2010, B3Yr was burned every three years during that period, and the control was never burned.

For the csv, the columns are as follows:
A - ignore
B - Year: year data was collected
C - Grass species
D - PixelCount: how many pixels were given the specific value for each species for each year
E - Percent: the percent of all of the pixels in the plot for that year that each species accounted for.

This data was used to create a sankey diagram.

Sharing/Access information
This is a section for linking to other ways to access the data, and for linking to sources the data is derived from, if any.

Links to other publicly accessible locations of the data:
YSE DROPBOX: https://www.dropbox.com/scl/fo/377vyg70k7tlli2qxv80d/AE-QXODej6dc-2rpWjRtp7g?rlkey=5g66o0rq9dfqe6521imb7vxn0&dl=0

Area1_Plots_GrassInventory.zip
GrassStack_20122025.tif
Metadata_variables_description.csv
composition_table_for_sankey.csv

README：1号区域草地覆盖度数据集
【请在此处插入Dryad平台上的数据集链接】

本文件夹内的数据为巴西马托格罗索州唐古罗野外站（Tanguro Field Station）1号区域的草地覆盖度时序栅格（raster）堆叠数据集。该数据集记录了2012年、2015年、2022年及2025年的草地覆盖情况。栅格数据由样地内草地覆盖度的网格化观测结果生成，栅格值依据植物物种进行赋值。本文件夹还包含一份仅针对火烧样地（B1yr、B3yr）的逗号分隔值（CSV）文件，其中记录了各年份不同物种的覆盖百分比。样地的矢量形状文件（shapefile）也已一并上传。原始数据由Divino Silverio及其唐古罗野外站团队采集。

### 数据与文件结构说明
栅格堆叠文件Grass_20122025.tif包含4个栅格波段，每个波段对应一个年份（2012年、2015年、2022年、2025年）。栅格值的含义如下：
0 = 林地（Forest）
1 = Gayanus须芒草（Andropogon gayanus）
2 = 长叶三芒草（Aristida longifolia）
3 = 混合草地（Grass-mix）
4 = 狼尾草属（Pennisetum sp.）
5 = 其他非禾本科植物（Other non-grass）
6 = 遮荫长叶三芒草*（Shaded Aristida longifolia）

*遮荫长叶三芒草指生长在林冠下的长叶三芒草，为2025年新增的分类类别。此前的调查未对林木覆盖下的草地进行量化统计。

关于形状文件（shapefile），其中的处理样地分为B1Yr、B3Yr与对照组（Control）。B1Yr样地在2004年至2010年间每年火烧一次；B3Yr样地在此期间每三年火烧一次；对照组则从未进行过火处理。

CSV文件的各列含义如下：
A列：忽略
B列：Year：数据采集年份
C列：Grass species：禾本科植物物种
D列：PixelCount：各年份对应物种的指定栅格值的像素数量
E列：Percent：该年份样地内各物种占总像素数的百分比

本数据集曾用于绘制桑基图（sankey diagram）。

### 共享与获取说明
本部分用于链接其他可获取该数据集的公开渠道，以及数据来源（如适用）。

该数据集的其他公开获取链接：
耶鲁环境学院Dropbox：https://www.dropbox.com/scl/fo/377vyg70k7tlli2qxv80d/AE-QXODej6dc-2rpWjRtp7g?rlkey=5g66o0rq9dfqe6521imb7vxn0&dl=0

包含的文件如下：
Area1_Plots_GrassInventory.zip、GrassStack_20122025.tif、Metadata_variables_description.csv、composition_table_for_sankey.csv