NASA Seasonal-to-Interannual Prediction Project Research Articles

Seasonal-to-Interannual Forecasting of Tropical Indian Ocean Sea Surface Temperature Anomalies: Potential Predictability and Barriers

Abstract Whether seasonally phased-locked persistence and predictability barriers, similar to the boreal spring barriers found for El Niño–Southern Oscillation (ENSO), exist for the tropical Indian Ocean sector climate is investigated using observations and hindcasts from two coupled ocean–atmosphere dynamical ensemble forecast systems: the National Centers for Environmental Prediction (NCEP) Coupled Forecast System (CFS) for 1990–2003, and the NASA Seasonal-to-Interannual Prediction Project (NSIPP) system for 1993–2002. The potential predictability of the climate is also assessed under the “perfect model/ensemble” assumption. Lagged correlations of the indices calculated over the east and west poles of the Indian Ocean dipole mode (IDM) index show weak sea surface temperature anomaly (SSTA) persistence barriers in boreal spring at both poles, but the major decline in correlation at the east pole occurs in boreal midwinter for all start months with an almost immediate recovery, albeit negative correlations, until summer approaches. Processes responsible for the change in sign of SSTAs associated with a major IDM event effect a similar change on much weaker SSTAs. At the west pole, a major decline occurs at the end of boreal summer for fall and winter starts when the thermocline deepens with the seasonal cycle and coupling between the ocean and atmosphere is weak. A decline in skillful prediction of SSTA at the east pole over boreal winter is also found in the hindcasts, but the relatively large thermocline depth anomalies are skillfully predicted through this time and skill in SSTA prediction returns. A predictability barrier at the onset of the boreal summer monsoon is found at both IDM poles with some return of skill in late fall. Potential predictability calculations suggest that this barrier may be overcome at the west pole with improvements to the forecast systems, but not at the east pole for forecasts initiated in boreal winter unless the ocean is initialized with a memory of fall conditions.

Characteristics of Diurnal and Seasonal Cycles in Global Monsoon Systems

In this paper, we characterize the climatological diurnal cycles over monsoon land and ocean regions, and seasonal cycles for six major monsoon systems around the world, from observations and outputs of the NASA Seasonal-to-Interannual Prediction Project (NSIPP) general circulation model outputs. Over monsoon land regions, the diurnal cycle has a pronounced late afternoon peak in rainfall and over the oceans a much weak diurnal signal with an early morning peak. The NSIPP model produces a daily peak in rainfall about 2–3 hours earlier than observed, possibly attributed to a cumulus parameterization which does not include cloud life cycle effects, and which lack detailed treatment of boundary layer processes. The seasonal cycles of all monsoon rain systems are controlled by an ITCZ and a subtropical and/or extratropical component due to the presence of contiguous large continental land mass. Strong climatological intraseasonal oscillations (CISO) are found in the monsoon systems of South Asia and East Asia, Australia, and South America, and to a lesser degree for West Africa and Mexico/North America. The NSIPP model captures the slow component seasonal component reasonably well, but the CISO’s are not well simulated. Analysis with reference to a new monsoon index shows that the major monsoon systems can be classified into three major groups (M1–3) according to the relative strength of continental vs. oceanic controls. East Asian and South Asian monsoons belong to a group (M1), where the continental influence is stronger than that from the ITCZ. The North America and South America monsoons show similar characteristics with the ITCZ influence slight stronger than continental influence (M2). The West Africa and Australia monsoons are found to be essentially ITCZ-monsoon systems (M3), with strong oceanic control, and limited poleward excursion of the monsoon rainbelt. We also find a clear trend associated with an increase in continentality for M3 monsoons, i.e., the West African and the Australian monsoons in the last two decades. We speculate that the former is linked to the partially recovery of the Sahel drought since the 1980’s.

Radiative and Dynamical Feedbacks over the Equatorial Cold Tongue: Results from Nine Atmospheric GCMs

Abstract The equatorial Pacific is a region with strong negative feedbacks. Yet coupled general circulation models (GCMs) have exhibited a propensity to develop a significant SST bias in that region, suggesting an unrealistic sensitivity in the coupled models to small energy flux errors that inevitably occur in the individual model components. Could this “hypersensitivity” exhibited in a coupled model be due to an underestimate of the strength of the negative feedbacks in this region? With this suspicion, the feedbacks in the equatorial Pacific in nine atmospheric GCMs (AGCMs) have been quantified using the interannual variations in that region and compared with the corresponding calculations from the observations. The nine AGCMs are the NCAR Community Climate Model version 1 (CAM1), the NCAR Community Climate Model version 2 (CAM2), the NCAR Community Climate Model version 3 (CAM3), the NCAR CAM3 at T85 resolution, the NASA Seasonal-to-Interannual Prediction Project (NSIPP) Atmospheric Model, the Hadley Centre Atmospheric Model (HadAM3), the Institut Pierre Simon Laplace (IPSL) model (LMDZ4), the Geophysical Fluid Dynamics Laboratory (GFDL) AM2p10, and the GFDL AM2p12. All the corresponding coupled runs of these nine AGCMs have an excessive cold tongue in the equatorial Pacific. The net atmospheric feedback over the equatorial Pacific in the two GFDL models is found to be comparable to the observed value. All other models are found to have a weaker negative net feedback from the atmosphere—a weaker regulating effect on the underlying SST than the real atmosphere. Except for the French (IPSL) model, a weaker negative feedback from the cloud albedo and a weaker negative feedback from the atmospheric transport are the two leading contributors to the weaker regulating effect from the atmosphere. The underestimate of the strength of the negative feedbacks by the models is apparently linked to an underestimate of the equatorial precipitation response. All models have a stronger water vapor feedback than that indicated in Earth Radiation Budget Experiment (ERBE) observations. These results confirm the suspicion that an underestimate of the regulatory effect from the atmosphere over the equatorial Pacific region is a prevalent problem. The results also suggest, however, that a weaker regulatory effect from the atmosphere is unlikely solely responsible for the hypersensitivity in all models. The need to validate the feedbacks from the ocean transport is therefore highlighted.

ENSO Bred Vectors in Coupled Ocean–Atmosphere General Circulation Models

Abstract The breeding method has been implemented in the NASA Seasonal-to-Interannual Prediction Project (NSIPP) coupled general circulation model (CGCM) with the ultimate goal of improving operational seasonal to interannual climate predictions through ensemble forecasting and data assimilation. This is the first attempt to isolate the evolving ENSO instability and its corresponding global atmospheric response in a fully coupled ocean–atmosphere GCM. The results herein show that the growth rate of the coupled bred vectors (BVs) is sensitive to the ENSO phases of the evolving background flow and peaks about 3 months before an ENSO event. The structure of the dominant growing BV modes also evolves with the background ENSO and exhibits a larger amplitude in the eastern tropical Pacific, reflecting the natural dynamical sensitivity associated with the shallow thermocline at the eastern boundary. The key features of coupled bred vectors of the NSIPP CGCM are reproduced when using the NCEP CGCM, an independently developed coupled general circulation model.

Seasonal Hydroclimate Variability over North America in Global and Regional Reanalyses and AMIP Simulations: Varied Representation

Abstract The monotony of seasonal variability is often compensated by the complexity of its spatial structure—the case in North American hydroclimate. The structure of hydroclimate variability is analyzed to provide insights into the functioning of the climate system and climate models. The consistency of hydroclimate representation in two global [40-yr ECMWF Re-Analysis (ERA-40) and NCEP] and one regional [North American Regional Reanalysis (NARR)] reanalysis is examined first, from analysis of precipitation, evaporation, surface air temperature (SAT), and moisture flux distributions. The intercomparisons benchmark the recently released NARR data and provide context for evaluation of the simulation potential of two state-of-the-art atmospheric models [NCAR's Community Atmospheric Model (CAM3.0) and NASA's Seasonal-to-Interannual Prediction Project (NSIPP) atmospheric model]. Intercomparisons paint a gloomy picture: great divergence in global reanalysis representations of precipitation, with the eastern United States being drier in ERA-40 and wetter in NCEP in the annual mean by up to a third in each case; model averages are like ERA-40. The annual means, in fact, mask even larger but offsetting seasonal departures. Analysis of moisture transport shows winter fluxes to be more consistently represented. Summer flux convergence over the Gulf Coast and Great Plains, however, differs considerably between global and regional reanalyses. Flux distributions help in understanding the choice of rainy season, especially the winter one in the Pacific Northwest; stationary fluxes are key. Land–ocean competition for convection is too intense in the models—so much so that the oceanic ITCZ in July is southward of its winter position in the both simulations! The overresponsiveness of land is also manifest in SAT; the winter-to-summer change over the Great Plains is 5–9 K larger than in observations, with implications for modeling of climate sensitivity. The nature of atmospheric water balance over the Great Plains is probed, despite unbalanced moisture budgets in reanalyses and model simulations. The imbalance is smaller in NARR but still unacceptably large, resulting from excessive evaporation in spring and summer. Adjusting evaporation during precipitation assimilation could lead to a more balanced budget. Global and regional reanalysis will remain of limited use for hydroclimate studies until they comply with the operative water and energy balance constraints.

Forecasting extreme events in the tropical Indian Ocean sector climate

Prospects for forecasting climate variability over the tropical Indian Ocean sector, specifically extreme positive events of the Indian Dipole Mode (IDM), with lead times of a season or more are investigated using the NASA Seasonal-to-Interannual Prediction Project (NSIPP) coupled-model system. The coupled system presents biases in its climatology over the Indian Ocean sector, which include (i) warmest sea-surface temperatures (SSTs) occurring in the central equatorial basin rather than on the eastside with the eastern (western) tropical SSTs up to 1 °C too cool (warm), (ii) a too northwest lying InterTropical Convergence Zonal over the ocean in boreal fall, (iii) a thermocline shallower (deeper) than observed west of Sumatra-Java (north of Madagascar), (iv) a delay of about a month in the onset (cessation) of the southwest (southeast) monsoon in the west (east) in boreal spring (fall). These biases affect the effectiveness of the SST-clouds-shortwave radiation negative feedback, the sensitivity of SST to wind-stress perturbations, and the character of equatorial coupled ocean-atmosphere modes. Despite these biases, ensemble hindcasts of the SST anomalies averaged over the eastern and western poles of the IDM for the decade 1993–2002, which included extreme positive events in 1994 and 1997/1998, are encouragingly good at 3-months lead. The onset of the 1997/1998-event is delayed by about a month, though the peak and decay are correctly timed. At 6-months lead-time, the forecast at the eastern pole deteriorates with either positive or negative false alarms generated each boreal fall. The forecast at the western pole remains good.

Climate Variability over the Tropical Indian Ocean Sector in the NSIPP Seasonal Forecast System

Abstract Prospects for forecasting Indian dipole mode (IDM) events with lead times of a season or more are examined using the NASA Seasonal-to-Interannual Prediction Project (NSIPP) coupled-model forecast system. The mean climatology of the system over the sector is reasonable, as determined from an almost-century-long run without data assimilation. However, the system presents biases, for example, too cool sea surface temperatures (SSTs), too shallow thermocline, and too strong southeasterlies along the Sumatra–Java coast in the east, and too warm SSTs, too deep thermocline, and too weak extension of the southeast trades into the Findlater jet in the west. These suggest coupling between the ocean and atmosphere is stronger, and the SST–clouds–shortwave radiation negative feedback less effective, than observed in the east with the opposite holding true in the west. Also, the negative zonal gradient in SST in the eastern equatorial basin, in contrast with the positive observed, suggests that equatorial Kelvin and Rossby coupled modes may have a different character from observed. Biases identified in the seasonal cycle, which may affect the strength and timing of IDM events, include a delayed onset of the boreal summer monsoon in the west, and a prolonged boreal summer monsoon in the east. Eight major positive IDM events occur during the almost-century-long run over a range of El Niño–Southern Oscillation phases with a tendency to occur post–El Niño/pre–La Niña. Consistent with the identified air–sea interaction biases, the cold (warm) anomaly at the east (west) pole tends to be stronger (weaker) than observed. Also, the cold anomaly extends much farther westward and is more equatorially trapped than observed; its slow westward propagation and the structure of the associated fields is reminiscent of an unstable, coupled Rossby mode with SST governed by lateral advection due to the westward displacement of the convective anomaly from the heat source. Otherwise, the life cycle of the eight-event composite is similar in seasonal phase locking and mechanisms of evolution and decay to the canonical event. For the decade from 1993 to the present, there were major positive IDM events in 1994 and 1997/98. Monthly mean SST anomalies over the western pole are well hindcast by the ocean component of the NSIPP system forced by observed surface fluxes with SST damped to observed values, and in which subsurface temperature data available in real time are assimilated; these data are very sparse over the Indian Ocean. Over the eastern pole, the SST anomalies are well hindcast except for the 1997/98 event, when it is too cool. The ensemble mean hindcast of the zonal surface wind anomaly of the central basin by the atmosphere component of the NSIPP system forced by observed SST is too weak during both events. These hindcasts provide initial conditions for the coupled system forecasts. Forecast ensembles for the decade 1993 onward, generated by the coupled system, give monthly mean SST anomalies averaged over the east and west poles of the IDM in agreement with observations at lead times of three months. The cool anomaly at the eastern pole is slightly too large in 1997/98, and the onset of the warm anomaly in 1997 is delayed by a month or so; its peak and decay are correctly timed. At lead times of six months, there is a significant deterioration in the forecast at the eastern pole with either false positive or negative alarms generated annually in boreal fall; that at the western pole remains good. These results are very encouraging and suggest that major IDM events have the potential to be forecast a season or more in advance.

A methodology for snow data assimilation in a land surface model

Snow cover has a large influence on heat fluxes between the land and atmosphere because of its high albedo and insulating thermal properties. Hence accurate snow representation in coupled land‐ocean‐atmosphere global climate models has the potential to greatly increase prediction accuracy. To this end, a one‐dimensional extended Kalman filter analysis scheme has been developed to assimilate observed snow water equivalent into the NASA Seasonal‐to‐Interannual Prediction Project (NSIPP) catchment‐based land surface model. This study presents the results from a set of data assimilation “twin” experiments using an uncoupled version of the land surface model. First, “true” snow states are generated by spinning‐up the land surface model for 1987 using an observation‐constrained version of the European Centre for Medium‐Range Weather Forecasts (ECMWF) 15‐year Re‐Analysis (ERA‐15) data set for atmospheric forcing. A degraded 1987 simulation was then made by initializing the model with no snow on 1 January 1987. A third simulation assimilated the synthetically generated snow water equivalent “observations” from the true simulation into the degraded simulation once a day. This study illustrates that by assimilating snow water equivalent observations, which are readily available from remote sensing satellites, other state variables (i.e., snow depth and temperature) can be retrieved and effects of poor initial conditions removed. Runoff and atmospheric flux predictions are also improved.

Intercomparison of Soil Moisture Memory in Two Land Surface Models

Abstract A heavy rain or a dry period can produce an anomaly in soil moisture, and the dissipation of this anomaly may take weeks to months. It is important to understand how land surface models (LSMs) used with atmospheric general circulation models simulate this soil moisture “memory,” because this memory may have profound implications for long-term weather prediction through land–atmosphere feedback. In order to understand better the effect of precipitation and net radiation on soil moisture memory, the NASA Seasonal-to-Interannual Prediction Project (NSIPP) Catchment LSM and the Mosaic LSM were both forced with a wide variety of idealized climates. The imposed climates had average monthly precipitation ranging from 15 to 500 mm and monthly net radiations (in terms of water equivalent) ranging from 20 to 400 mm, with consequent changes in near-surface temperature and humidity. For an equivalent water holding capacity, the two models maximize memory in distinctly different climate regimes. Memory in the...

Impact of Land Surface Initialization on Seasonal Precipitation and Temperature Prediction

Abstract The potential role of land initialization in seasonal forecasting is illustrated through ensembles of simulations with the NASA Seasonal-to-Interannual Prediction Project (NSIPP) model. For each boreal summer during 1997–2001, two 16-member ensembles of 3-month simulations were generated. The first, “AMIP style” (Atmospheric Model Intercomparison Project) ensemble establishes the degree to which a perfect prediction of SSTs would contribute to the seasonal prediction of precipitation and temperature over continents. The second ensemble is identical to the first, except that the land surface is also initialized with “realistic” soil moisture contents through the continuous prior application (within GCM simulations leading up to the start of the forecast period) of a daily observational precipitation dataset and the associated avoidance of model drift through the scaling of all surface prognostic variables. A comparison of the two ensembles shows that land initialization has a statistically signifi...

Influence of the Interannual Variability of Vegetation on the Surface Energy Balance—A Global Sensitivity Study

Abstract The degree to which the interannual variability of vegetation phenology affects hydrological fluxes over land is investigated through a series of simulations with the Mosaic land surface model, run both offline and coupled to the NASA Seasonal-to-Interannual Prediction Project (NSIPP) atmospheric general circulation model (GCM). Over a 9-yr period, from 1982 to 1990, interannual variations of global biophysical land surface parameters (i.e., vegetation density and greenness fraction) are derived from Normalized Difference Vegetation Index (NDVI) data collected by the Advanced Very High Resolution Radiometers (AVHRRs). First the sensitivity of evapotranspiration to interannual variations in vegetation properties is evaluated through offline simulations that ignore feedbacks between the land surface and the atmospheric models, and interannual precipitation variations. Evapotranspiration is shown to be highly sensitive to variations in vegetation over wet continental surfaces that are not densely ve...

A methodology for initializing soil moisture in a global climate model: Assimilation of near‐surface soil moisture observations

Because of its long‐term persistence, accurate initialization of land surface soil moisture in fully coupled global climate models has the potential to greatly increase the accuracy of climatological and hydrological prediction. To improve the initialization of soil moisture in the NASA Seasonal‐to‐Interannual Prediction Project (NSIPP), a one‐dimensional Kalman filter has been developed to assimilate near‐surface soil moisture observations into the catchment‐based land surface model used by NSIPP. A set of numerical experiments was performed using an uncoupled version of the NSIPP land surface model to evaluate the assimilation procedure. In this study, “true” land surface data were generated by spinning‐up the land surface model for 1987 using the International Satellite Land Surface Climatology Project (ISLSCP) forcing data sets. A degraded simulation was made for 1987 by setting the initial soil moisture prognostic variables to arbitrarily wet values uniformly throughout North America. The final simulation run assimilated the synthetically generated near‐surface soil moisture “observations” from the true simulation into the degraded simulation once every 3 days. This study has illustrated that by assimilating near‐surface soil moisture observations, as would be available from a remote sensing satellite, errors in forecast soil moisture profiles as a result of poor initialization may be removed and the resulting predictions of runoff and evapotranspiration improved. After only 1 month of assimilation the root‐mean‐square error in the profile storage of soil moisture was reduced to 3% vol/vol, while after 12 months of assimilation, the root‐mean‐square error in the profile storage was as low as 1% vol/vol.

The Impact of Detailed Snow Physics on the Simulation of Snow Cover and Subsurface Thermodynamics at Continental Scales

The three-layer snow model is coupled to the global catchment-based Land Surface Model (LSM) of the NASA Seasonal to Interannual Prediction Project (NSIPP) project, and the combined models are used to simulate the growth and ablation of snow cover over the North American continent for the period 1987-1988. The various snow processes included in the three-layer model, such as snow melting and re-freezing, dynamic changes in snow density, and snow insulating properties, are shown (through a comparison with the corresponding simulation using a much simpler snow model) to lead to an improved simulation of ground thermodynamics on the continental scale.