Tuesday, April 21, 2015
Augmenting rainfall probability to ward off long-term drought?
Despite the ridiculous pseudo controversy surrounding global warming in the public discourse, the reality is that global warming is real and has already significantly started influencing the global climate. One of the most important factors in judging the range and impact of global warming as well as how society should respond is also one of the more perplexing, cloud formation. Not only do clouds influence the cycle of heat escape and retention, but they also drive precipitation probability. Precipitation plays an important role in maintaining effective hydrological cycles as well as heat budgets and will experience significant changes in reaction to future warming largely producing more extreme outcomes with some areas receiving significant increases that will produce flash flooding whereas other areas will be deprived of rainfall producing longer-term droughts similar to those now seen in California.
At its core precipitation is influenced by numerous factors like solar heating and terrestrial radiation.1,2 Of these factors various aerosol particles are thought to hold an important influence. Both organic and inorganic aerosols are plentiful in the atmosphere helping to cool the surface of Earth by sunlight scattering or serving as nuclei support for the formation of water droplets and ice crystals.3 Not surprisingly information regarding the means in which the properties of these aerosols influence cloud formation and precipitation is still limited, which creates significant uncertainties in climate modeling and planning. Therefore, increasing knowledge of how aerosols influence precipitation will provide valuable information for managing the various changes that will occur and even possibly mitigating those changes.
The formation of precipitation within clouds is heavily influenced by ice nucleation. Ice nucleation involves the induction of crystallization in supercooled water (supercooled = a meta-stable state where water is in liquid form at below typical freezing temperatures). The process of ice nucleation typically occurs through one of two pathways: homogenous or heterogeneous. Homogeneous nucleation entails spontaneous nucleation within a properly cooled solution (usually a supersaturated solution of relative humidity of 150-180% with a temperature of around –38 degrees C) requiring only liquid water or aqueous solution droplets.4-6 Due to its relative simplicity homogeneous nucleation is better understood than heterogeneous nucleation. However, because of the temperature requirements homogeneous nucleation typically only takes place in the upper troposphere and with a warming atmosphere it should be expected that its probability of occurrence would reduce.
Heterogeneous nucleation is more complicated because of the multiple pathways that can be taken, i.e. depositional freezing, condensation, contact, and immersion freezing.7,8 Typically these different pathways allow for more flexibility in nucleation with generic initiation conditions beginning at just south of 0 degrees C and a relative humidity of 100%. This higher temperature fails to prevent nucleation because of the presence of a catalyst, a non-water based substance that is commonly referred to as an ice-forming nuclei (IN). Also heterogeneous nucleation can involve diffusive growth in a mixed-phase cloud that consumes liquid droplets at a faster rate (Wegener–Bergeron–Findeisen process) than super-cooled droplets or snow/graupel aggregation.9
Laboratory experiments have demonstrated support for many different materials acting as IN: different metallic particles, biological materials, certain glasses, mineral dust, anhydrous salts, etc.8,10,11 These laboratory experiments involve wind tunnels, electrodynamic levitation, scanning calorimetry, cloud chambers, and optical microscopy.12,13 However, not surprisingly there appears a significant difference between nucleation ability in the lab and in nature.8,10
Also while homogenous ice nucleation is exactly that, heterogeneous nucleation does not have the same quenching properties.8 Temperature variations within a cloud can produce differing methods of heterogeneous nucleation versus homogeneous nucleation producing significant differences in efficiency. For example not surprisingly some forms of nucleation in cloud formations are more difficult to understand like high concentration formation in warm precipitating cumulus clouds; i.e. particle concentrations increasing from 0.01 L-1 to 100 L-1 in a few minutes at temperatures exceeding –10 degrees C and outpacing existing ice nucleus measurements.14 One explanation for this phenomenon is the Hallett-Mossop (H-M) method. This method is thought to achieve this rapid freezing through interaction with a narrow band of supercooled raindrops producing rimers.15
The H-M methodology requires cloud temperatures between approximately –1 and –10 degrees C with the availability of large rain droplets (diameters > 24 um), but at a 0.1 ratio relative to smaller (< 13 um droplets).16,17 When the riming process begins ice splinters are ejected and grow through water vapor deposition producing a positive feedback effect increasing riming and producing more ice splinters. Basically a feedback loop develops between ice splinter formation and small drop freezing. Unfortunately there are some questions whether or not this methodology can properly explain the characteristics of secondary ice particles and the formation of ice crystal bursts under certain time constraints.18 However, these concerns may not be accurate due to improper assumptions regarding how water droplets form relative to existing water concentrations.15
One of the more important element of rain formation in warm precipitating cumulus clouds, in addition to other cloud formations, appears to involve the location of ice particle concentrations at the top of the cloud formation where there is a higher probability for large droplet formation (500 – 2000 um diameters).15 In this regard cloud depth/area is a more important influencing element than cloud temperature.19 In addition the apparent continued formation of ice crystals stemming from the top proceeding downwards can produce raindrop freezing that catalyzes ice formation creating a positive feedback and ice bursts.20
This process suggests that there is a sufficient replenishment of small droplets at the cloud top increasing the probability of sufficient riming. It is thought that the time variation governing the rate of ice multiplication and how cloud temperature changes accordingly is determined by dry adiabatic cooling at the cloud top, condensational warming, evaporational cooling at the cloud bottom.15 Bacteria also appear to play a meaningful role in both nucleating primary ice crystals and scavenging secondary crystals.7 Even if bacteria concentrations are low (< 0.05 L-1) the catalytic effect of nucleating bacteria produces a much more “H-M” friendly environment.
The most prominent inorganic aerosol that acts as an IN is dust commonly from deserts that is pushed into the upper atmosphere by storms.21,22 The principal origin of this dust is from the Sahara Desert, which is lofted year round versus dust from other origin points like the Gobi or Siberia. While the ability of this dust to produce rain is powerful it can also have a counteracting effect as a cloud condensation nuclei (CCN). In most situations when CCN concentration is increased raindrop conversion becomes less efficient, especially for low-level clouds (in part due to higher temperatures) largely by reducing riming efficiency.
The probability of dust acting as a CCN is influenced by the presence of anthropogenic pollution, which typically is a CCN on its own.23,24 In some situations the presence of pollution could also increase the overall rate of rainfall as it can suppress premature rainfall allowing more rain droplets to crystallize increasing riming and potential rainfall. However, this aspect of pollution is only valid in the presence of dust or other INs for if there is a dearth of IN concentration, localized pollution will decrease precipitation.25 Soot can also influence nucleation and resultant rainfall, but only under certain circumstances. For example if the surface of the soot contains available molecules to form hydrogen bonds (typically from available hydroxyl and carbonyl groups) with available liquid water molecules nucleation is enhanced.26 Overall it seems appropriate to label dust as a strong IN and anthropogenic pollution as a significant CCN.
In mineral collection studies and global simulations of aerosol particle concentrations both deposition and immersion heterogeneous nucleation appear dominated by dust concentrations acting as INs, especially in cirrus clouds.10,27,28 Aerosols also modify certain cloud properties like droplet size and water phase. Most other inorganic atmospheric aerosols behave like cloud condensation nuclei (CCN), which assist the condensation of water vapor for the formation of cloud droplets in a certain level of super-saturation.25 Typically this condensation produces a large number of small droplets, which can reduce the probability of warm rain (above freezing point).29,30
Recall that altitude is important in precipitation, thus it is not surprising that one of the key factors in how aerosols influence precipitation type and probability appears to involve the elevation and temperature at which they interact. For example in mixed-phase clouds, the top area increases relative to increases in CCN concentrations versus a smaller change at lower altitudes and no changes in pure liquid clouds.15,31 Also CCN only significantly influence temperatures when top and base cloud temperatures are below freezing.31 In short it appears that CCN influence is reduced relative to IN influence at higher altitudes and lower temperatures.
Also cloud drop concentration and size distribution at the base and top of a cloud determine the efficiency of the CCN and are dictated by the chemical structure and size of an aerosol. For example larger aerosols have a higher probability of becoming CCN over IN due to their coarse structure. Finally and not surprisingly overall precipitation frequency increases with high water content and decreases with low water content when exposed to CCNs.31 This behavior creates a positive feedback structure that increases aerosol concentration, so for arid regions the probability of drought increases and in wet regions the probability of flooding increases.
While dust from natural sources as well as general pollution are the two most common aerosols, an interesting secondary source may be soil dust produced from land use due to deforestation or large-scale construction projects.32-34 These actions create anthropogenic dust emissions that can catalyze a feedback loop that can produce greater precipitation extremes; thus in certain developing economic regions that may be struggling with droughts continued construction in effort to improve the economy could exacerbate droughts. Therefore, developing regions may need to produce specific methodologies to govern their development to ensure proper levels of rainfall for the future.
While the role of dust has not been fully identified on a mechanistic level, its importance is not debatable. The role of biological particles, like bacteria, is more controversial and could be critical to identifying a method to enhance rainfall probability. It is important to identify the capacity of bacteria to catalyze rainfall for some laboratory studies have demonstrated that inorganic INs only have significant activity below –15 degrees C.10,35 For example in samples of snowfall collected globally originating at temperatures of –7 degrees C or warmer a vast majority of the active IN, up to 85%, were lysozyme-sensitive (i.e. probably bacteria).36,37 Also rain tends to have higher proportions of active IN bacteria than air in the same region.38 With further global warming on the horizon air temperatures will continue to increase lowering the probability window for inorganic IN activity, thus lowering the probability of rainfall in general (not considering any other changes born from global warming).
Laboratory and field studies have demonstrated approximately twelve species of bacteria with significant IN ability spread within three orders of the gammaproteobacteria with the two most notable/frequent agents being Pseudomonas syringae and P. fluorescens and to a lesser extent Xanthomonas.39,40 In the presence of an IN bacterium nucleation can occur at temperatures as warm as –1.5 degrees C to –2 degrees C.41,42 These bacteria appear to have the ability to act as IN due to the existence of a single gene that codes for a specific membrane protein that catalyzes crystal formation by acting as a template for water molecule arrangement.43 The natural origins of these bacteria derive mostly from surface vegetation.
Supporting the idea of the key membrane scaffolding, an acidic pH environment can significantly reduce the effectiveness of bacteria-based nucleation.45,46 Also these protein complexes for nucleation are larger for warmer temperature nucleating bacteria, thus more prone to breakdown in higher acidic environments.44,46 Therefore, low lying areas that have significant acidic pollution like sulfurs could see a reduction in precipitation probability over time. Also it seems that this protein complex could be the critical element to bacteria-based nucleation versus the actual biological processes of the bacteria as nucleation was augmented even when the bacteria itself was no longer viable.46
Despite laboratory and theoretical evidence supporting the role of bacteria in precipitation, as stated above what occurs in the laboratory serves little purpose if it does not translate to nature. This translation is where a controversy arises. It can be difficult to separate the various particles within clouds from residue collection due to widespread internal mixing, but empirical evidence demonstrates the presence of biological material in orographic clouds.47 Also ice nucleation bacteria are present over all continents as well as in various specific locations like the Amazon basin.37,48,49
Some estimates have suggested that 10^24 bacteria enter the atmosphere each year and stay circulating between 2 and 10 days allowing bacteria, theoretically, to travel thousands of miles.50,51 However, there is a lack of evidence for bacteria in the upper troposphere and their concentrations are dramatically lower than those of inorganic materials like dust and soot.28,35,52 Based on this lack of concentrations questions exist to the efficiency of how these bacteria are aerosolized over their atmospheric lifetimes. One study suggests that IN active bacteria are much more efficiently precipitated than non-active IN bacteria, which may explain the disparity between the observations in the air, clouds and precipitation.53
Another possible explanation for this disparity is that most biological particles are generated on the surface and are carried by updrafts and currents into the atmosphere. While the methods of transport are similar to inorganic particles, biological particles have a higher removal potential due to dry or wet deposition due to their typical greater size. Therefore, from a nature standpoint bacteria reside in orographic clouds because they are able to participate in their formations, but are not able to reach higher cloud formations, so most upper troposphere rain is born from dust not bacteria.
Some individuals feel that the current drop freezing assays, which are used to identify the types of bacteria and other agents in a collected sample, can be improved upon to produce a higher level of discrimination between the various classes of IN active bacteria that may be present in the sample. One possible idea is to store the sample at low temperatures and observe the growth and the type of IN bacteria that occur in a community versus individual samples.54 Perhaps new identification techniques would increase the ability to discern the role of bacteria in cloud formation and precipitation.
Among the other atmospheric agents and their potential influence on precipitation potassium appears to have a meaningful role. Some biogenic emissions of potassium, especially around the Amazon, can act as catalysts for the beginning process of organic material condensation.55 However, this role seems to ebb as potassium mass fraction drops as the condensation rate increases.55 This secondary role of potassium as well as the role of bacteria may signal an important element to why past cloud seeding experiments have not achieve the hypothesized expectations.
The lack of natural bacteria input into higher cloud formations leads to an interesting question. What would happen if IN active bacteria like P. syringae were released via plane or other increased altitude method that would result in a higher concentration of bacteria in these higher altitude cloud formations? While typical cloud formation involves vapor saturation due to air cooling and/or increased vapor concentration, increased IN active bacteria concentration could also speed cloud formation as well as precipitation probability.
Interestingly in past cloud seeding experiments orographic clouds appear to be more sensitive to purposeful seeding versus other cloud formations largely because of the shorter residence times of cloud droplets.56,57 One of the positive elements of seeding appears to be that increased precipitation in the target area does not reduce the level of precipitation in surrounding areas including those beyond the target area. In fact it appears that there is a net increase (5-15%) among all areas regardless of the location of seeding.58 The previous presumption that there was loss appears to be based on randomized and not properly controlled seeding experiments.58
The idea of introducing increased concentrations of IN active bacteria is an interesting one if it can increase the probability of precipitation. Of course possible negatives must be considered for such an introduction. The chief negative that could be associated with such an increase from a bacterium like P. syringae would be the possibility of more infection of certain types of plants. The frost mechanism of P. syringae is a minor concern because most of the seeding would be carried out between late spring and early fall where night-time temperatures should not be cold enough to induce freezing. Sabotaging the type III secretion system in P. syringe via some form of genetic manipulation should reduce, if not eliminate, the plant invasion potential. Obviously controlled laboratory tests should be conducted to ensure a high probability of invasion neutralization success before any controlled and limited field tests are conducted. If the use of living bacteria proves to be too costly, exploration of simply using the key specific membrane protein is another possible avenue of study.
Overall the simple fact is that due to global warming, global precipitation patterns will change dramatically. The forerunner to these changes can already been seen in the state of California with no reasonable expectation for new significant levels of rainfall in sight. While other potable water options are available like desalinization, the level of infrastructure required to divert these new sources from origins source to usage points will be costly and these processes do have significant detrimental byproducts. If precipitation probabilities can be safely increased through new cloud seeding strategies like the inclusion of IN active bacteria it could go a long way to combating some of the negative effects of global warming while the causes of global warming itself are mitigated.
1. Zuberi, B, et Al. “Heterogeneous nucleation of ice in (NH4)2SO4-H2O particles with mineral dust immersions.” Geophys. Res. Lett. 2002. 29(10). 1504.
2. Hung, H, Malinowski, A, and Martin, S. “Kinetics of heterogeneous ice nucleation on the surfaces of mineral dust cores inserted into aqueous ammonium sulfate particles.” J. Phys. Chem. 2003. 107(9):1296-1306.
3. Lohmann, U. “Aerosol effects on clouds and climate.” Space Sci. Rev. 2006. 125:129-137.
4. Hartmann, S, et Al. “Homogeneous and heterogeneous ice nucleation at LACIS: operating principle and theoretical studies.” Atmos. Chem. Phys. 2011. 11:1753-1767.
5. Cantrell, W, and Heymsfield, A. “Production of ice in tropospheric clouds. A review.” American Meteorological Society. 2005. 86(6):795-807.
6. Riechers, B, et Al. “The homogeneous ice nucleation rate of water droplets produced in a microfluidic device and the role of temperature uncertainty.” Physical Chemistry Chemical Physics. 2013. 15(16):5873-5887.
7. Cziczo, D, et Al. “Clarifying the dominant sources and mechanisms of cirrus cloud formation.” Science. 2013. 340(6138):1320-1324.
8. Pruppacher, H, and Klett, J. “Microphysics of clouds and precipitation.” (Kluwer Academic, Dordrecht. Ed. 2, 1997). pp. 309-354.
9. Lance, S, et Al. “Cloud condensation nuclei as a modulator of ice processes in Arctic mixed-phase clouds.” Atmos. Chem. Phys. 2011. 11:8003-8015.
10. Hoose, C, and Mohler, O. “Heterogeneous ice nucleation on atmospheric aerosols: a review of results from laboratory experiments.” Atmos. Chem. Phys. 2012. 12:9817-9854.
11. Abbatt, J, et Al. “Solid ammonium sulfate aerosols as ice nuclei: A pathway for cirrus cloud formation.” Science. 2006. 313:1770-1773.
12. Murray, B, et Al. “Kinetics of the homogeneous freezing of water.” Phys. Chem. 2010. 12:10380-10387.
13. Chang, H, et Al. “Phase transitions in emulsified HNO3/H2O and HNO3/H2SO4/H2O solutions.” J. Phys. Chem. 1999. 103:2673-2679.
14. Hobbs, P, and Rangno, A. “Rapid development of ice particle concentrations in small, polar maritime cumuliform clouds.” J. Atmos. Sci. 1990. 47:2710–2722.
15. Sun, J, et Al. “Mystery of ice multiplication in warm-based precipitating shallow cumulus clouds.” Geophysical Research Letters. 2010. 37:L10802.
16. Hallett, J, and Mossop, S. “Production of secondary ice particles during the riming process.” Nature. 1974. 249:26-28.
17. Mossop, S. “Secondary ice particle production during rime growth: The effect of drop size distribution and rimer velocity.” Q. J. R. Meteorol. Soc. 1985. 111:1113-3324.
18. Mason, B. “The rapid glaciation of slightly supercooled cumulus clouds.” Q. J. R. Meteorol. Soc. 1996. 122:357-365.
19. Rangno, A, and Hobbs, P. “Microstructures and precipitation development in cumulus and small cumulous-nimbus clouds over the warm pool of the tropical Pacific Ocean. Q. J. R. Meteorol. Soc. 2005. 131:639-673.
20. Phillips, V, et Al. “The glaciation of a cumulus cloud over New Mexico.” Q. J. R. Meteorol. Soc. 2001. 127:1513-1534.
21. Karydis, V, et Al. “On the effect of dust particles on global cloud condensation nuclei and cloud droplet number.” J. Geophys. Res. 2011. 166:D23204.
22. Connolly, P, et Al. “Studies of heterogeneous freezing by three different desert dust samples.” Atmos. Chem. Phys. 2009. 9:2805-2824.
23. Lynn, B, et Al. “Effects of aerosols on precipitation from orographic clouds.” J. Geophys. Res. 2007. 112:D10225.
24. Jirak, I, and Cotton, W. “Effect of air pollution on precipitation along the Front Range of the Rocky Mountain.” J. Appl. Meteor. Climatol. 2006. 45:236-245.
25. Fan, J, et Al. “Aerosol impacts on California winter clouds and precipitation during CalWater 2011: local pollution versus long-range transported dust.” Atmos. Chem. Phys. 2014. 14:81-101.
26. Gorbunov, B, et Al. “Ice nucleation on soot particles.” J. Aerosol Sci. 2001. 32(2):199-215.
27. Kirkevag, A, et Al. “Aerosol-climate interactions in the Norwegian Earth System Model – NorESM. Geosci. Model Dev. 2013. 6:207-244.
28. Hoose, C, Kristjansson, J, Burrows, S. “How important is biological ice nucleation in clouds on a global scale?” Environ. Res. Lett. 2010. 5:024009.
29. Lohmann, U. “A glaciation indirect aerosol effect caused by soot aerosols.” Geophys. Res. Lett. 2002. 29:11.1-4.
30. Koop, T, et Al. “Water activity as the determinant for homogeneous ice nucleation in aqueous solutions.” Nature. 406:611-614.
31. Li, Z, et Al. “Long-term impacts of aerosols on the vertical development of clouds and precipitation.” Nature Geoscience. 2011. DOI: 10.1038/NGEO1313
32. Zender, C, Miller, R, and Tegen, I. “Quantifying mineral dust mass budgets: Terminology, constraints, and current estimates.” Eos. Trans. Am. Geophys. Union. 2004. 85:509-512.
33. Forester, P, et Al. “Changes in atmospheric constituents and in radiative forcing. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
34. O’Sullivan, D, et Al. “Ice nucleation by fertile soil dusts: relative importance of mineral and biogenic components.” Atmos. Chem. Phys. 2014. 14:1853-1867.
35. Murray, B, et Al. “Ice nucleation by particles immersed in supercooled cloud droplets.” Chem. Soc. Rev. 2012. 41:6519-6554.
36. Christner, B, et Al. “Geographic, seasonal, and precipitation chemistry influence on the abundance and activity of biological ice nucleators in rain and snow. PNAS. 2008. 105:18854. dio:10.1073/pnas.0809816105.
37. Christener, B, et Al. “Ubiquity of biological ice nucleators in snowfall.” Science. 2008. 319:1214.
38. Stephanie, D, and Waturangi, D. “Distribution of ice nucleation-active (INA) bacteria from rainwater and air, NAYATI Journal of Biosciences. 2011. 18:108-112.
39. Vaitilingom, M, et Al. “Long-term features of cloud microbiology at the puy de Dome (France). Atmos. Environ. 2012. 56:88-100.
40. Cochet, N and Widehem, P. “Ice crystallization by Pseudomonas syringae.” Appl. Microbiol. Biotechnol. 2000. 54:153-161.
41. Heymsfield, A, et Al. “Upper-tropospheric relative humidity observations and implications for cirrus ice nucleation.” Geophys. Res. Lett. 1998. 25:1343-1346.
42. Twohy, C, and Poellot, M. “Chemical characteristics of ice residual nuclei in anvil cirrus clouds: implications for ice formation processes.” Atmos. Chem. Phys. 2005. 5:2289-2297.
43. Joly, M, et Al. “Ice nucleation activity of bacteria isolated from cloud water.” Atmos. Environ. 2013. 70:392-400.
44. Attard, E, et Al. “Effects of atmospheric conditions on ice nucleation activity of Pseudomonas.” Atmos. Chem. Phys. 2012. 12:10667-10677.
45. Kawahara, H, Tanaka, Y, and Obata H. “Isolation and characterization of a novel ice-nucleating bacterium, Pseudomonas, which has stable activity in acidic solution.” Biosci. Biotechnol. Biochem. 1995. 59:1528-1532.
46. Kozloff, L, Turner, M, and Arellano, F. “Formation of bacterial membrane ice-nucleating lipoglycoprotein complexes.” J. Bacteriol. 1991. 173:6528-6536.
47. Pratt, K, et Al. “In-situ detection of biological particles in high altitude dust-influenced ice clouds.” Nature Geoscience. 2009. 2:dio:10.1038/ngeo521.
48. Prenni, A, et Al. “Relative roles of biogenic emissions and Saharan dust as ice nuclei in the Amazon basin.” Nat. Geosci. 2009. 2:402-405.
49. Phillips, V, et Al. “Potential impacts from biological aerosols on ensembles of continental clouds simulated numerically.” Biogeosciences. 2009. 6:987-1014.
50. Burrows, S, et Al. “Bacteria in the global atmosphere – Part 1: review and synthesis of literature data for different ecosystems.” Atmos. Chem. Phys. 2009. 9:9263-9280.
51. Burrows, S, et Al. “Bacteria in the global atmosphere – Part 2: modeling of emissions and transport between different econsystems.” Atmos. Chem. Phys. 2009. 9:9281-9297.
52. Despres, V, et Al. “Primary biological aerosol particles in the atmosphere: a review. Tellus B. 2012. 64:349-384.
53. Amato, P, et Al. “Survival and ice nucleation activity of bacteria as aerosols in a cloud simulation chamber.” Atmos. Chem. Phys. Discuss. 2015. 15:4055-4082.
54. Stopelli, E, et Al. “Freezing nucleation apparatus puts new slant on study of biological ice nucleators in precipitation.” Atmos. Meas. Tech. 2014. 7:129-134.
55. Pohlker, C, et Al. “Biogenic potassium salt particles as seeds for secondary organic aerosol in the Amazon.” Science. 2012. 337(31):1075-1078.
56. Givati, A, and Rosenfeld, D. “Separation between cloud-seeding and air-pollution effects.” J. Appl.Meteorol. 2005. 44:1298-1314.
57. Givati, A, et Al. “The Precipitation Enhancement Project: Israel - 4 Experiment. The
Water Authority, State of Israel. 2013. pp. 55.
58. DeFelice, T, et Al. “Extra area effects of cloud seeding – An updated assessment.” Atmospheric Research. 2014. 135-136:193-203.