With leading climate change scientists concluding that “It is very likely that heat waves will occur with a higher frequency and duration” as the global mean temperatures increases , it seems likely that tightening bottlenecks, rather than changes in mean temperatures, will be responsible for temperature related damages in agriculture for the next few decades. We note that few adaptation techniques seem to address them directly. In this paper, we assess the potential benefits of a novel adaptation technique, targeted at a rapidly developing climate bottleneck for California pistachio. Warming winter days, with temperatures far below the yearly maximum but high enough to disrupt the normal tree metabolism, could harm output to a great extent. This bottleneck stands out in the existing literature in two ways: first, while much of the climate change in agriculture literature focuses on annual crops, pistachio trees are perennial. Moreover, the bottleneck occurs not in the “growing season”, but on the winter months when trees are dormant. Second, while many researched bottlenecks are temperature distribution right side tail effects our bottleneck is not at the extreme yearly temperatures. Rather, higher winter day time temperatures cause a decline in a special temperature metric known as chill portions. Weakly correlated with the mean yearly temperature, this change is easy to be missed by researchers focusing on traditional climate metrics. As we will see, this non-intuitive change in climate, at a non-intuitive season to explore,square plastic pot could have effects in the billions of dollars for pistachio. Scientists at the University of California Cooperative Extension have been experimenting with a new technique for dealing with the risk of insufficient winter chill, caused by raised winter daytime temperatures.

The treatment consists of covering the dormant pistachio trees with a non-toxic chemical mix, which blocks the sunlight and prevents overheating of tree buds. This acts as a de-facto shading device, lowering daytime temperatures from December to February. Similar solutions, such as installing nets for partial shading in orchards, are being explored in other fruit crops. The literature on adaptation considers many options, including adoption of existing technologies and innovations. Among the innovations, the emphasis has been on irrigation technologies, diversification in varieties, and weather information . The concept of MCE has not been explored, even though the technologies used for this purpose need not be completely new to farmers. Unrelated to climate change, existing MCE solutions for frost can be witnessed in Northern California’s vineyards. Frosts are bottlenecks: they last a few hours and create great damages to crops. Many growers are equipped with wind machines and propane gas heaters, preventing frost damage by circulating warm air around the vines on frost nights. A similar MCE solution for frost is implemented by some almond growers in California, who reportedly hire helicopters to fly over their orchards . Almonds and wine being the first and fourth top agricultural exports of California , the benefits of MCE in these crops are probably very high. Archaeologists have identified implementation of MCE by the Tiwanako civilization in the 7-12 centuries. On the shores of lake Titikaka, a cold and arid area, frosts are common. To deal with this , the Tiwanako raised their fields artificially with sediment. The fields were irrigated by canals carrying water diverted from nearby springs, and the moist soil served as a heat storage unit that protected crops during frosts . At the height of the Tiwanako civilazation, this practice supported a population estimated 10 times larger than the area’s population in the 1990’s . We offer the first documentation of a MCE implementation for climate change adaptation, and believe this concept could be substantial in its economic analysis. In some cases, techniques for dealing with new or enhanced climate bottlenecks, targeting them on the field level, could delay the eventual crop transition for decades.

Our case study shows substantial benefits from MCE for one crop in California, yet similar style solutions might be developed and used for other crops in other parts of the world. The remaining of this paper is organized as following: Section 2 borrows from the pest control literature to create a micro-economic model for MCE application. In section 3, we adapt this model to California pistachio, and present the functional forms and parameters used for simulations. Section 4 presents the simulation results, and section 5 concludes. Introduced to California more than 80 years ago, and grown commercially since the mid 1970’s, pistachio was the state’s 9th leading agricultural product in gross value in 2014, generating a total revenue of $1.65 Billion. California produces virtually all the pistachio crop in the US, and competes internationally with Iran and Turkey . In 2014, five California counties were responsible for a 96.2% of the state’s pistachio crop: Kern , Fresno , Tulare , Madera and Kings . Since the year 2000, total harvested acres in these counties have been increasing by roughly 10% yearly. Each increase represent a 6 – 7 year old investment decision, as trees need to mature before commercial harvest . Like many other fruit and nut trees, pistachio requires a minimal input of winter chill, a temperature metric measured in portions. This brief explanation of chill is based on Erez .This phase is an evolutionary adaptation, allowing the tree to “hibernate” and protect sensitive organs while harsh weather conditions take place. Trees prepare for dormancy by storing energy reserves, shedding leaves, and developing organs to protect the meristems. Once a tree went into dormancy, it needs to calculate when to optimally “wake up”. Blooming too early might expose the foliage to frost. Blooming too late means not taking advantage of available resources , and eventually being out-competed. Temperatures and day lengths affect both entry and exit from dormancy. Agronomists stipulate that, once dormant, tree buds “count” chill portions and measure day lengths, until threshold levels of both are reached. Only then will the buds break and the tree will start blooming. Failure to attain a threshold chill count, varying between crops and cultivars, leads to low and non-uniform bud breaking, which is linked to low yields at harvest.

Thus chill accumulation is critical for growers, especially in warmer areas where the chill constraint might be binding. To accumulate a chill portion, temperatures need to remain below certain levels for a few days. Roughly speaking, when temperatures go above 6oC, accumulation slows down. When temperatures exceed 15oC, the count reverses, quickly rounding down to the last integer portion that has been “banked”. Thus, rising winter daytime temperatures can have a detrimental effect on chill count,square plant pot even if the temperatures themselves are not extreme on the yearly distribution. Agronomists estimate the minimum requirement for the common pistachio cultivars in California at 54 – 58 portions. Compared to other popular fruit and nut crops in the state, this is a high threshold , putting pistachio on the verge of not attaining its chill requirements in some California counties. In fact, there is evidence of low chill already hurting yields . Chill in most of California is predicted to decline in the next decades. Luedeling, Zhang, and Girvetz estimate the potential chill drop for the southern part of San Joaquin valley, where over 96% of California pistachio is currently grown. For the measure of first decile, i.e. the amount of portions fulfilled in 90% of years, they predict a drop from an estimate of 64.3 chill portions in the year 2000 to estimates ranging between 50.6 and 54.5  in the years 2045-2060. Scientists at the University of California Cooperative Extension have been experimenting with potential solutions for the threat of low chill. One solution, tested successfully on a small scale experiments, involves spraying a non-toxic chemical mix, based on kaolin clay, on the dormant pistachio trees. Acting as a de-facto shading device, this creates a special microclimate for the chill counting tree buds. Shaded from winter sunlight, they now experience lower effective daytime temperatures, which raises their count of chill portions. In experiments, the chill portion count on treated trees was higher than on untreated trees; and the treated trees produced more pistachio clusters . Kaolin itself is already used by growers for other purposes, and research continues on the potential of other commonly used reflecting substances for this task . With relatively cheap application costs, this technique can help raise the chill counts in orchards, and at times save harvests. To assess the potential gains from this new technology, we follow Zilberman et al. and Hueth, Cohen, and Zilberman to create simulation outcome distributions by randomly drawing parameter values from pre-determined ranges and solving the model in equation . Parameters includes the climate prediction distribution, elasticities of supply and demand, and a range of MCE prices. The parameter ranges and distributions are specified in subsections below. To calculate the gains from MCE in our simulations, outcomes need to be compared to a benchmark or counter-factual, a world where MCE does not exist.

To do this, we run the simulations again, but restricting xi = 0 for all counties. This results in the no MCE market outcomes, where the weather bottleneck damages are not abated. Comparing simulations with and without MCE, yet identical in all other parameters, we can calculate the profit gains , consumer surplus gains, and total welfare gains from MCE in our model. We abstract away from a benchmark with increased storage, which could alleviate inter- year fluctuations, for two main reasons. First, pistachio is usually stored for up to one year . Second, the alternate bearing nature of pistachio means a great amount of storage is already in place. In 2005, the stock to output ratio in California pistachio was 0.37 . Given the limited storage time and already existing supply smoothening storage operations, it seems unlikely that extra storage would serve as a meaningful solution to the new threat of insufficient chill. We cannot use the above-mention chill predictions by Luedeling, Zhang, and Girvetz , as they uniformly cover all of our counties and use a first decile statistic which we find less useful. We therefore present our county specific calculations for current and future predicted chill at each county. A winter’s chill portion count is calculated from a vector of hourly temperatures. Observed temperatures for 2007-2016 come from the California Irrigation Management Information System , a network of weather stations located in many counties in California, operated by the California Department of Water Resources. Given that goal of CIMIS is to provide decision making data for farmers, stations are located in rural areas. We use the location of stations to set the reference point at each of the five counties: the representative point at each county is the centroid of active CIMIS stations at that county. This seems better than using the counties’ geographic centroid, as many of them extend east of the San Joaquin Valley into the less cultivated Sierra. To estimate future chill, we use temperature predictions of a CCSM4 model from CEDA . These predictions use an RCP8.5 scenario. This scenario assumes a global mean surface temperature increase of 2o C between 2046-2065 . The predicted temperatures for the years 2025-2040 are produced for each county. Following Leard and Roth , we perform quantile calibration on the 2007-2016 past predictions, which can be compared with the actual observed temperatures2 . Figure 2 shows a map of the counties, reference points, CIMIS stations, and interpolation points. Once we have the observed temperatures from CIMIS and the predicted future temperatures, we calculate the past and predicted chill count for each winter at each county, as specified in Erez and Fishman . Figure 3 compares the distributions of observed chill and predicted chill . In three out of the five pistachio producing counties, the chill distribution is predicted to lower in such manner, that low chill years would be quite common. Madera and Tulare seem not to suffer an adverse change in chill counts. More details on the climate data processing are found in Appendix B. We test this 5-county chill prediction matrix in 2025 – 2050 for a the null hypothesis of multivariate-normally distributed chill predictions, and fail to reject it.