This study investigated the relationship between the share of agriculture in GDP and some socio-economic and climate variables using ARMA, ARDL and structural change estimation in R. The data, covering the period 1960-2016, were obtained from various secondary sources, including CBN, WDI, FAOSTAT and NiMet. The results indicated that Nigerian population is on a very rapid rise while GDP is showing a decline. The trend of the other variables included shows slow decline or rise. The forecast to 2030 and beyond shows that only population and aquaculture will continue to rise. However, lagged values of agricultural GDP, CO2 emission and the difference between maximum and minimum temperature significantly and positively affected GDP at 1% level while differences between maximum and minimum temperature significantly and negatively affected GDP at 5% level. In the short-term, population and L(coint) are negatively significant (p=0) while the difference between maximum and minimum temperature negatively affected share of agriculture in GDP at 1% level of significance. In the long-term, CO2 emission and population positively (p=0) affected the share of agriculture in GDP while arable land, cereal yield and aquaculture negatively (p=0) affected the share of agriculture in GDP determining four structural breaks in GDP since 1960. Although, the results implied mixed effects of climate change, it is important for Nigeria to critically look at the frequent changes in policy, negative effects of aquaculture on GDP which might be as a result of shift of mobile resources from main sector as well as the involvement of non-experts in formulating agricultural policies for Nigeria.