Prominent experiments have shown that species richness increases the stability of aboveground net primary productivity (ANPP) in synthetic grasslands.  But in most cases, species richness also increases ANPP, so it is unclear if the stability of ANPP responds to species richness or pre-disturbance levels of ANPP.  To compare species richness and pre-disturbance ANPP as predictors of stability of ANPP, I experimentally disturbed two ecologically restored plant communities where species richness and ANPP were not initially correlated.  In 2008 and 2009, I sampled species richness and ANPP across a range of species richness within a Wisconsin prairie restoration and a California salt marsh restoration.  I treated my own shoot biomass removal as a disturbance, and considered 2008 data pre-disturbance and 2009 data post-disturbance.  I calculated three commonly used indicators of stability: recovery (2009 ANPP), resilience (2008 ANPP – 2009 ANPP), and proportional resilience (2009 ANPP/2008 ANPP).

Results/Conclusions

Results from the prairie and salt marsh plant communities were similar.  Species richness was not significantly related to recovery, resilience, or proportional resilience in either site.  Pre-disturbance ANPP was positively correlated with recovery of ANPP in the prairie (p=0.0101, R²=0.11, n=60) and the salt marsh (p=0.0041, R²=0.18, n=45).  Pre-disturbance ANPP was positively related to resilience in both sites (p<0.0001, R²>0.40), somewhat negatively related to proportional resilience in the prairie (p=0.0923, R²=0.05, n=60), and negatively related to proportional resilience in the salt marsh (p=0.0001, R²=0.30, n=45).  The resilience and proportional resilience results likely represent an artifact since high-shoot-biomass are inherently able to lose more ANPP and might be less able to regain it all in one year. 

Pre-disturbance ANPP, rather than species richness, appears to control the recovery of ANPP in these restored plant communities.  My data corroborate recent findings by other investigators working in experimental grasslands.  The finding that post-disturbance ANPP is mostly determined by the pre-disturbance ANPP suggests that: (1) intrinsic features and conditions of the sites (nutrient levels, light, etc.) set bounds on ANPP in a given area and/or (2) legacy effects of the pre-disturbance community (propagules of former species and/or resprouting from root stores) allow vary with pre-disturbance ANPP.  My data suggest both factors were at play in the two restorations.  The factors controlling recover of ANPP should be examined in a wide variety of contexts, and in the absence of confounding relationships between species richness and ANPP.