Obviously agro-ecosystem out of all types terrestrial ecosystem...

This is a wide open question, and it seems like there's no intuitive answer, at least from my tree-centric perspective. You might expect xeric systems to be resistant to drought because they have drought-tolerant (drought-adapted) species. Then again they are stressful environments even in a good year, and plants might be near their threshold for aridity. A mesic site might be more buffered in water availability. But mesic and moist systems support species that might be very drought sensitive. Furthermore the plants might have high transpirational demands due to the biomass they built up during favorable years. At a stand scale, it might be that plants grows until most of the water capacity is used, so that similar water deficits are felt across a range of precipitation. A lot of species are growing near their hydraulic safety margin regardless of site moisture (Choat et al. 2012), so maybe all these factors tend to cancel out.

So there's no clear global answer as of yet, but maybe some system-specific answers. It seems like in dry coniferous forests, dry sites are more drought-prone, maybe because of limited NSCs to defend against herbivory. But in mesic deciduous forests, it might be the wetter sites that are more drought-prone. But it's probably species specific and depends upon what other species are growing in the stand. Jim Clark has some good recent work in this arena. Greenwood et al. 2017 try to tackle this question and ended up not finding a difference between biomes, but it might be a process that operates within a biome (e.g. at the hillslope scale). They did, however, find some associations with functional traits, which might give us a generalizable species-based approach -- the answer to the question you asked might be an interaction between drought, stand structure, site conditions, and species traits that alter their drought resistance. I think this is a really interesting, and important question, but not one with a good answer yet.

There's plenty of good papers on the topic, despite the lack of consensus/synthesis yet. If you haven't read it, I particularly enjoyed Jump et al. 2017, which has a good review and some interesting ideas.

Choat et al. 2012. Global convergence in the vulnerability of forests to drought. Nature.

Clark et al. 2016. The impacts of increasing drought on forest dynamics,

structure, and biodiversity in the United States. Global Change Biology.

Clark et al. 2014. Competition-interaction landscapes for the joint response of forests to climate change. Global Change Biology.

Greenwood Et al. 2017. Tree mortality across biomes is promoted by drought intensity, lower wood density and higher specific leaf area. Ecology Letters.

Jump et al. 2017 Structural overshoot of tree growth with climate variability and the global spectrum of drought-induced forest dieback. Global Change Biology.

I think grass land ecossytem is higly sensitive to drought.

Thanks "Samuel Flake", "Anoop Kumar Srivastava" and "Sangamreddi Chandramouli" for your answer.

Maybe surprisingly? Wetlands are sensitive to drought events. Even intertidal systems (seagrasses and salt marshes) that are flushed with with tidal water twice a day can be pushed to collapse due to drought events. See e.g. our papers:

Article Drought, Mutualism Breakdown, and Landscape-Scale Degradatio...

Article The role of changing climate in driving the shift from peren...

I hope this helps?

Regards, Jim

Please have a look at very informative study.....

Contrasting responses of water use efficiency to drought across global terrestrial ecosystems( Scientific Reports 6, Article number: 23284 (2016),doi:10.1038/srep23284)

Abstract: Drought is an intermittent disturbance of the water cycle that profoundly affects the terrestrial carbon cycle. However, the response of the coupled water and carbon cycles to drought and the underlying mechanisms remain unclear. Here we provide the first global synthesis of the drought effect on ecosystem water use efficiency (WUE = gross primary production (GPP)/evapotranspiration (ET)). Using two observational WUE datasets (i.e., eddy-covariance measurements at 95 sites (526 site-years) and global gridded diagnostic modelling based on existing observation and a data-adaptive machine learning approach), we find a contrasting response of WUE to drought between arid (WUE increases with drought) and semi-arid/sub-humid ecosystems (WUE decreases with drought), which is attributed to different sensitivities of ecosystem processes to changes in hydro-climatic conditions. WUE variability in arid ecosystems is primarily controlled by physical processes (i.e., evaporation), whereas WUE variability in semi-arid/sub-humid regions is mostly regulated by biological processes (i.e., assimilation). We also find that shifts in hydro-climatic conditions over years would intensify the drought effect on WUE. Our findings suggest that future drought events, when coupled with an increase in climate variability, will bring further threats to semi-arid/sub-humid ecosystems and potentially result in biome reorganization, starting with low-productivity and high water-sensitivity grassland.

I think all ecosystems are sensitive to drought. However, organismal living in xeric environments developed several adaptation mechanisms that allow them to better cope with these drought episodes (e.g. phenotipic plasticity). Functional responses to drought is then different among humid and arid ecosystems as revealed the recent paper of Yang et al. (2016-Nature) provided above. It must have important effects on net ecosystem exchange in the future global climatic change. Rainfall pulse effects on ecosystem processes in xeric environments are well studied but drought effects on semi-arid and humid ecosystems not so much. It is an interesting topic to be developed in the future.