What question did this study set out to answer?

This research aims to establish a regression equation to predict NH3 generation based on meteorological conditions at livestock farms.

May 9, 2026Open Access

A study on derivation of meteorological data regression equation for prediction of livestock derived NH3

Key Points

This research aims to establish a regression equation to predict NH3 generation based on meteorological conditions at livestock farms.
Installed NH3 sensors at 8 farms in Eumseong-gun, Korea for 6 months.
Collected real-time NH3, temperature, humidity, and rainfall data in 30-minute increments.
Performed Pearson's correlation and multiple regression analysis to derive relationships and regression equations.
Pearson's coefficients indicated a strong correlation between NH3 and environmental conditions, ranging from -0.738 to 0.839.
Monthly regression equations were derived, but R² values were below 0.65 for cross-month data validation.

Abstract

This study examined the causal relationship between environmental conditions and NH 3 generation by installing NH 3 sensors at 8 livestock farms in Eumseong-gun, Korea.NH 3 and meteorological data were collected for 6 months (September 2021 to February 2022) through the livestock odor management system, and the amount of NH 3 measured in real time and the temperature, humidity, and rainfall data measured at the same time through sensors in the field were pretreated in 30-minute increments.To identify the correlation between NH 3 and environmental conditions, Pearson's coefficient and Variance Inflation Factors (VIF) analysis were performed to derive a positive/negative linear relationship, and the time range was selected as 4 months from September 2021 to December 2021.In addition, one location was selected from each of the four farmhouses as a spatial range.The correlation between NH 3 , temperature, humidity, and rainfall showed a high Pearson's coefficients ranging from -0.738 to 0.839.Monthly regression equations were derived through multiple regression analysis, and as a result of calculating R 2 by substituting data from other months, it was found that it was not high below 0.65.

Demander à l'IA

Bookmark

View Full Paper