This README.txt file was generated on <2023-12-06> by <Sergio Morell-Monzó>

-------------------
GENERAL INFORMATION
-------------------

Title of Dataset: <Georeferenced dataset of abandoned citrus crops (Comunitat Valenciana, Spain)>

Author Information:

	Principal Investigator: <Sergio Morell-Monzó. Universitat Politécnica de València. EPS Gandia. C/ Paranimf, 1. 46730 Gandia, València (España). sermomon@upv.es. ORCID: 0000-0001-8883-2618>

	Asociate or Co-investigator: <María-Teresa Sebastiá-Frasquet. Departamento de Ingeniería Hidráulica y Medio Ambiente. Universitat Politécnica de València EPS Gandia. C/ Paranimf, 1. 46730 Gandia, València (España). mtsebastia@hma.upv.es. ORCID: 0000-0002-8042-5628>

	Asociate or Co-investigator: <Javier Estornell Cremades. Departamento de Ingeniería Cartográfica, Geodésica y Fotogrametría. Universitat Politécnica de València EPS Gandia. C/ Paranimf, 1. 46730 Gandia, València (España). jaescre@cgf.upv.es. ORCID: 0000-0003-0854-5358>

Date of data collection: <2019, 2020, 2021>

Geographic location of data collection: <Comunitat Valenciana (Spain). [38.936752, -0.160847]>

Information about funding sources or sponsorship that supported the collection of the data: <This data collection was partially created thanks to the funds provided by the Regional Government of Spain, Generalitat Valenciana, within the framework of the research project AICO/2020/246.>

General description: <This dataset contains georeferenced agricultural citrus parcels with information on crop status (non-productive, productive or abandoned) for the years 2019, 2020 and 2021 in GeoJSON format. The dataset is divided into 7 subsets (OLV_19, LSF_20, LSF_21, PV_21, TV_21, BP_21, and NL_21) in different locations of the Comunitat Valenciana (Spain). However, most of the data are located in the region of La Safor. Each of these subsets has its own characteristics in terms of sampling/classification method, date of acquisition and number of parcels. In total there are 1676 agricultural parcels labeled. The boundaries of the parcels correspond to those of the Geographic Information System of Agricultural Parcels (SIGPAC) database, which corresponds to the cadastral database (“parcela” in SIGPAC nomenclature). Each parcel may contain one or more subplots ("recinto" in SIGPAC nomenclature) but with only one crop status. Data are derived from: Morell Monzó, S. (2023). “Desarrollo de procedimientos para la detección del abandono de cultivos de cítricos utilizando técnicas de teledetección” [Tesis doctoral]. Universitat Politècnica de València.  https://doi.org/10.4995/Thesis/10251/193058>

Keywords: <land use / land cover classification, agriculture, crop monitoring, agricultural land abandonment, remote sensing, Earth observation, ground truth>

--------------------------
SHARING/ACCESS INFORMATION
--------------------------

Open Access to data: <Open>

Licenses/restrictions placed on the data, or limitations of reuse: <Creative Commons BY-NC-SA>

Citation for and links to publications that cite or use the data:

Morell-Monzó, S.; Estornell, J.; Sebastiá-Frasquet, M.-T. Comparison of Sentinel-2 and High-Resolution Imagery for Mapping Land Abandonment in Fragmented Areas. Remote Sensing, 2020, 12, 2062. https://doi.org/10.3390/rs12122062

Morell-Monzó, S.; Sebastiá-Frasquet, M.-T.; Estornell, J. Land Use Classification of VHR Images for Mapping Small-Sized Abandoned Citrus Plots by Using Spectral and Textural Information. Remote Sensing, 2021, 13, 681. https://doi.org/10.3390/rs13040681

Morell-Monzó, S.; Sebastiá-Frasquet, M.-T.; Estornell, J. Cartografía Del Abandono de Cultivos de Cítricos Mediante El Uso de Datos Altimétricos: LiDAR y Fotogrametría SfM. Revista de Teledetección, 2022, 47–58. https://doi.org/10.4995/raet.2022.16698

Morell-Monzó, S; Garófano-Gómez, V. Investigación del abandono de tierras en la comarca de La Safor (Comunitat Valenciana) utilizando datos Sentinel-2. Confederación de Centros para el Desarrollo Rural: Centre de Desenvolupament Rural La Safor (COCEDER - CDR La Safor), Octubre, 2022

Morell-Monzó, S.; Estornell, J.; Sebastiá-Frasquet, M.-T. Assessing the Capabilities of High-Resolution Spectral, Altimetric, and Textural Descriptors for Mapping the Status of Citrus Parcels. Computers and Electronics in Agriculture, 2023, 204, 107504. https://doi.org/10.1016/j.compag.2022.107504

Morell-Monzó, S.; Sebastiá-Frasquet, M.-T.; Estornell, J.; Moltó, E. Detecting Abandoned Citrus Crops Using Sentinel-2 Time Series. A Case Study in the Comunitat Valenciana Region (Spain). ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 201, 54–66. https://doi.org/10.1016/j.isprsjprs.2023.05.003

--------------------
DATA & FILE OVERVIEW
--------------------

File list:

	-> Summary.xlsx

	-> Data.zip

		- OLV_19_x240_14-05-2019_14_07_2019_v1.1.geojson
		- LSF_20_x280_03-02-2020_06-05-2020_v1_1.geojson
		- LSF_21_x836_09-03-2021_16-05-2021_v1_1.geojson
		- PV_21_x080_16-05-2021_15-05-2021_v1.1.geojson
		- TV_21_x080_15-05-2021_15-05-2021_v1.1.geojson
		- BP_21_x080_13-05-2021_13-05-2021_v1.1.geojson
		- NL_21_x080_22-07-2021_22-07-2021_v1.1.geojson

	-> GEE_assets.js

Type of version of the dataset: <1.1>

Total size: <487 KB>

--------------------------
METHODOLOGICAL INFORMATION
--------------------------

Description of methods used for collection/generation of data:

	- OLV_19 dataset (OLV_19_x0240_14-05-2019_14_07_2019_v1.1.geojson) consists of 240 parcels selected through a stratified random sampling and classified during field visits carried out between July 11 and 14, 2019 in the municipality of Oliva. The class identified during the field visits was verified using the 2019 orthophotos of the Institut Cartogràfic Valencià.

	- LSF_20 dataset (LSF_20_x280_03-02-2020_06-05-2020_v1_1.geojson) LS_20 consists of 280 parcels selected through a systematic and intensive sampling of 12 areas surveyed by an Unmanned Aerial Vehicle (UAV) in La Safor region and classified by visual interpretation of the photos taken by the UAV photogrammetric flight. The class identified with the UAV was verified using the 2020 orthophotos of the Institut Cartogràfic Valencià.

	- LSF_21 dataset (LSF_21_x836_09-03-2021_16-05-2021_v1_1.geojson) consists of 240 parcels selected through a stratified random sampling and classified during field visits carried out between July 11 and 14, 2019 in the coastal area between Gandia and Oliva (La Safor region). The class identified during the field visits was verified using the 2021 orthophotos of the Institut Cartogràfic Valencià.

	- PV_21 (PV_21_x080_16-05-2021_16-05-2021_v1_1.geojson) consists of 80 parcels selected through a random sampling over a 20km x 6km in the municipalities of Potries and Villalonga (La Safor region) and classified by photo-interpretation of the 2019, 2020, 2021 orthophotos of the Institut Cartogràfic Valencià and Street View images from Google Maps.

	- TV_21 (TV_21_x080_15-05-2021_15-05-2021_v1_1.geojson) consists of 80 parcels selected through a random sampling over a 20km x 6km in the municipality of Tavernes de la Valldigna (La Safor region) and classified by photo-interpretation of the 2019, 2020, 2021 orthophotos of the Institut Cartogràfic Valencià and Street View images from Google Maps.

	- BP_21 (BP_21_x080_13-05-2021_13-05-2021_v1_1.geojson) consists of 80 parcels selected through a random sampling over a 20km x 6km in the municipalities of Benicull and Polinyà del Xuquer (Ribera Baixa region) and classified by photo-interpretation of the 2019, 2020, 2021 orthophotos of the Institut Cartogràfic Valencià and Street View images from Google Maps.

	- NL_21 (NL_21_x080_22-07-2021_22-07-2021_v1_1.geojson) consists of 80 parcels selected through a random sampling over a 20km x 6km in the municipality of Nules (Plana Baixa region) and classified by photo-interpretation of the 2019, 2020, 2021 orthophotos of the Institut Cartogràfic Valencià and Street View images from Google Maps.

Methods for processing the data: <The description of the data processing is described in Morell-Monzó, S.; Sebastiá-Frasquet, M.-T.; Estornell, J.; Moltó, E. Detecting Abandoned Citrus Crops Using Sentinel-2 Time Series. A Case Study in the Comunitat Valenciana Region (Spain). ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 201, 54–66. https://doi.org10.1016j.isprsjprs.2023.05.003.>

Software- or Instrument-specific information needed to interpret the data, including software and hardware version numbers: <QGIS 3.34. The dataset can also be accessed through Google Earth Engine using the GEE_asset.js file>

Describe any quality-assurance procedures performed on the data: <The data were collected through field campaigns and an expert technician verified the status of the crop using RGB orthophotos with a spatial resolution of 0.25 m from the Institut Cartogràfic Valencià.>

--------------------------
DATA-SPECIFIC INFORMATION
--------------------------

The Summary.xlsx file contains a summary of the data content.

Coordinate Reference System: ETRS89_UTM_30N (EPSG:25830)

Projection: Transverse Mercator

Variables: 24

Variable names: ["PARCEL_ID", "DATASET", "CLASS_NUM", "CLASS_CODE", "CLASS_NAME", "CLASS_DATE", "VERIF_DATE", "YEAR", "AREA_HA", "PERIM_M", "SAMP_METHO", "CLASS_METH", "CENTR_LON", "CENTR_LAT", "CROSSVAL_S", "id", "dn_oid", "provincia", "municipio", "zona", "poligono", "parcela", "agregado", "referencia"]

Variables abbreviations:

	- PARCEL_ID: Parcel identifier within the subset. From 1 to n (n = number of parcels in the subset).
	- DATASET: Dataset code.
	- CLASS_NUM: Numeric code defining the crop status (1: non-productive, 2: productive, 3: abandoned).
	- CLASS_CODE: Abbreviation of the crop status (NP: non productive, PR: productive, AB: abandoned).
	- CLASS_NAME: Crop status.
	- CLASS_DATE: Date of crop classification in the field. If NA the field was only classified by photointerpretation.
	- VERIF_DATE: Date of verification of the crop status on the Institut Catogràfic Valencià orthophoto (date on which the flight was taken).
	- YEAR: Year of dataset (not agricultural season).
	- AREA_HA: Parcel area in hectares.
	- PERIM_M: Perimeter of the parcel in meters.
	- SAMP_METHO: Method used for selecting the parcels (sampling method).
	- CLASS_METH: Method used for classification (class assignment).
	- CENTR_LON: Longitude of the parcel centroid.
	- CENTR_LAT: Latitude of the parcel centroid.
	- CROSSVAL_S: 4-folds cross-validation split identifier. Divide the dataset into 4 random splits with proportional number of parcels of each class in all groups.
	- id: id. From SIGPAC joining.
	- dn_oid: From SIGPAC.
	- provincia: From SIGPAC.
	- municipio: From SIGPAC.
	- zona: From SIGPAC.
	- poligono: From SIGPAC.
	- parcela: From SIGPAC.
	- agregado: From SIGPAC.
	- referencia: From SIGPAC.
	
Number of cases/rows: 1676

Legend (CLASS_CODE / CLASS_NUM / CLASS_NAME):

	- NP [CLASS_NAME == "non_productive"]: Land use/cover which includes bare soil orchards before tree planting and recently planted trees that have not reached productive age and are predominantly occupied by bare or low vegetated soil.

	- PR [CLASS_NAME == "productive"]: Land use/cover which includes orchards at productive age (> 5 years). There are no signs of abandonment and the trees are in good health.

	- AB [CLASS_NAME == "abandoned"]: Land use/cover which includes abandoned orchards with physically evident signs of abandonment like: excessive growth of wild vegetation, removal or loss of vigor of trees.

Version history:

	- v1.0: Preliminar version (author version).
	- v1.1: Some variables available in version 1.0, which contains appreciations made during field visits, has been removed. Some parcel boundaries are also revised to keep consistency with SIGPAC.