A Dataset on Nelore Cattle Supplementation in the Midwest Region of Brazil for Integrated Analysis of Supplementation, Satellite Remote Sensing, and Weather Variables
Description
This is a dataset of an experiment performed in in the School Farm of the Federal University of Mato Grosso do Sul (UFMS). UFMS is in the Midwest region of Brazil (Mato Grosso do Sul State). The school farm is located at 20°26'37.7"S 54°50'58.5"W. The data was collected in a 16ha paddock containing 26 Nelore breed animals with brachiaria Decumbens forage. The dataset contains weight and supplementation data for each animal, Multispectral data, and environmental data for a period of one year (December, 2022 - October, 2023). Along the period, every 13-28 days, all 26 animals in the paddock got their weights registered, as well as their supplementation data (supplementation delivered (kg) per day, average daily weight gain, time of each supplementation, and data for how many times and for how long each animal went to the feeder). For each days-interval, dates (START_DATE and FINAL_DATE), period identification (PERIOD) and day range (AMOUNT_DAYS) were registered, as well as animal identification (ANIMAL), its weight at the start and end (START_WEIGHT and FINAL_WEIGHT), its average daily weight gain (ADG), average daily supplementation (ADS), time for each daily supplementation (SUP_TIME), total supplementation amount for the whole herd (SUP_TOTAL), total time the animal spent at the feeder (TOTAL_TIME), how many times the animal went to the feeder (TOTAL_ATTENDANCE) and total of different days the animal spent at the feeder (TOTAL_DAYS). Specifically, a couple more specific filter were applied to attendance and days data in order to get more specific supplementation data for each animal. We split the data for specific ranges of time ((6-12h), (12-18h), (18-00h), (6-8h), (12-14h), (18-20h)) and for specific date ranges (spring (09/23-12/21), summer (12/22-03/20), autumn (03/23-06/21), winter (06/22-09/22)) Multispectral data was acquired looking at the paddock location. The data was collected using Google Earth Engine API, based on Sentinel-2 multispectral images. Twenty sentinel-2 bands (BAND1, BAND2, BAND3, BAND4, BAND5, BAND6, BAND7, BAND8A, BAND9, BAND11, and BAND12) were acquired, and eight spectral indices (NDVI_INDEX, NDWI_INDEX, EVI_INDEX, LAI_INDEX, DVI_INDEX, GCI_INDEX, GEMI_INDEX, and SAVI_INDEX) were calculated and integrated into the dataset. Weather data was also collected using the Open-Meteo weather API. The data acquired are Mean Temperature of the period (TEMP_AVG), Rain sum registered in the period (RAIN_SUM), Average daily rain registered during the period (RAIN_AVG), Solar Radiation sum and Average Solar Radiation registered during the period (RAD_SOL_SUM,RAD_SOL_AVG), Average evapotranspiration registered during the period (EVAPOT), Average Relative Humidity registered during the period (HUM_REL_AVG) and Average Atmospheric Pressure registered during the period (PRES_ATM_AVG).
Files
Steps to reproduce
The complete dataset with cattle supplementation, weight, multispectral and weather data is in the "Complete_Dataset.csv" file. The “Table_1_Supplementation_Data.csv” file contains cattle weight and supplementation data. The “Table_2_Field_Multispectral_Data.csv” file contains multispectral data of the paddock. The "Table_3_Weather_Data.csv" file contains weather data of the paddock. If you want to extract out multispectral and weather data using the API, please follow the next steps: 1. On-site https://developers.google.com/earth-engine/guides/service_account create your private key to access the google earth engine API. 2. Insert your keys into “get_image_data.ipynb” and run it. The script will generate a file called "Table_2_Field_Multispectral_Data.csv" containing the desired data. After obtaining the multispectral data, run "collect_weather_data.ipynb" with "Table_2_Field_Multispectral_Data.csv" and "Table_1_Supplementation_Data.csv" in the same folder to obtain the "Complete_Dataset.csv" file.