TITLE:
CO2 Forcing of Changes in Lower Tropospheric Temperatures: A Time Series Analysis
AUTHORS:
Peter R. Hartley
KEYWORDS:
Satellite-Measured Temperature Anomalies, Forcing-Feedback Climate Model, Stationarity, Autoregressive Integrated Moving-Average (ARIMA), Autoregressive Fractionally Integrated Moving-Average (ARFIMA), Vector Autoregression (VAR)
JOURNAL NAME:
American Journal of Climate Change,
Vol.14 No.4,
December
22,
2025
ABSTRACT: A simple forcing-feedback model of the effect of CO2 on temperatures implies that the natural logarithm of the concentration of CO2 in the atmosphere,
lnC
, and satellite-measured lower tropospheric temperatures,
T
, should have the same order of integration. Multiple tests provide strong evidence that
lnC
has a unit root (it is integrated of order 1). However, the hypothesis that
T
has a unit root is strongly rejected both for the globe as a whole and for a set of non-overlapping geographic subsets. A possible explanation is that CO2 accumulation is associated with some other trending process that moderates its effects on atmospheric temperatures in such a way as to prevent
T
from also having a unit root. The analysis also reveals interactions between temperature anomalies in the non-overlapping geographic subsets that further limit the range of models consistent with the evidence.