Co-developing seamless climate information for the wine sector

Spring frost events occurring after budburst of grapevines can damage new shoots, disrupt plant growth and cause large economic losses to the viticultural sector. Frost protection practices encompass a variety of vineyard management actions across timescales, from seasonal to decadal and beyond. The cost-effectiveness of such measures depends on the availability of accurate predictions of the relevant climate hazards at the appropriate timescales.In this work, we present a statistical downscaling method which predicts variations in the frequency of occurrence of spring frost events in the important winemaking region of Catalunya at the seasonal timescale. The downscaling method exploits the seasonal predictability associated with the predictable components of the atmospheric variability over the Euro-Atlantic region, and produces local predictions of frost occurrence at a spatial scale relevant to vineyard management.The downscaling method is designed to address the specific needs highlighted by a representative stakeholder in the local viticultural sector, and is expected to deliver an actionable prototype climate service. The statistical procedure is developed in perfect prognosis mode: the method is trained with large-scale reanalysis data against a high-resolution gridded observational reference, and validated against multi-model seasonal hindcast predictions. Our work spotlights the potential benefits of transferring climate predictability across spatial scales for the design and provision of usable climate information, particularly regarding extremes.

<p>There is an increasing demand from society and policy makers for reliable, robust and actionable climate information for the upcoming 40 years. However, providing such seamless information poses a challenge to the scientific community. Traditionally, the scientific community developed targeted forecasts for specific time scales, e.g. short-term, seasonal or decadal predictions. These model integrations have thus a limited forecast period and do not provide seamless information on time scales up to 40 years.</p><p>This work discusses two alternative approaches to combine information from initialized decadal predictions (providing information up to ten years) with uninitialized climate projections (available until 2100 and beyond). </p><p>The first is  a novel framework, which is designed to implicitly make use of the (added) values from initialization by constraining uninitialized climate projections using decadal predictions. This approach is applied to near-surface temperatures over the North Atlantic Subpolar gyre region from CMIP5 models. Results suggest that such a constraining approach is able to provide more skillful, seamless climate information beyond decadal time-scales compared to using unconstrained climate projections. </p><p>The second approach is based on the simple temporal concatenation of decadal predictions and climate projections. It is shown that this can introduce inconsistencies, which may impact the usability for potential end users. Two different methods to overcome these issues are discussed: the application of a simple calibration method and a weighting scheme based on model performance. Results for the calibration method are in general promising, whereas the impact of the model weighting scheme is smaller. The latter is mainly associated with the small size of the decadal prediction ensemble, which hinders the usual application of the weighting scheme as done in previous studies based on much larger ensembles.</p>

Climate variability and change, associated with increasing water demands, can have significant implications for water availability. In the Brazilian semi-arid, eutrophication in reservoirs raises the risk of water scarcity. The reservoirs have also a high seasonal and annual variability of water level and volume, which can have important effects on chlorophyll-a concentration (Chla). Assessing the influence of climate and hydrological variability on phytoplankton growth can be important to find strategies to achieve water security in tropical regions with similar problems. This study explores the potential of machine learning models to predict Chla in reservoirs and to understand their relationship with hydrological and climate variables. The model is based mainly on satellite data, which makes the methodology useful for data-scarce regions. Tree-based ensemble methods had the best performances among six machine learning methods and one parametric model. This performance can be considered satisfactory as classical empirical relationships between Chla and phosphorus may not hold for tropical reservoirs. Water volume and the mix-layer depth are inversely related to Chla, while mean surface temperature, water level, and surface solar radiation have direct relationships with Chla. These findings provide insights on how seasonal climate prediction and reservoir operation might influence water quality in regions supplied by superficial reservoirs.

The impact of soil moisture anomalies on the seasonal mean atmospheric predictability over North America is investigated on the basis of a 100‐year simulation by an atmospheric general circulation model (AGCM). It is shown that soil moisture anomalies have the smallest persistence during the late spring and summer months, yet the associated near‐surface atmospheric climate anomalies are the largest. The causes for this seasonality are traced to the seasonal variation of the surface evaporation and the modulating control of the atmospheric dynamic variability. It is also shown that the dominant spatial modes of interannual soil moisture variability in late spring can have different degrees of interaction with the surface climate. To what extent do initial soil moisture anomalies and their subsequent evolution impact the atmospheric variability? This question is of relevance if the soil moisture anomalies are to be observed and are used as one of the initial conditions in seasonal climate forecast. To investigate this, we carried out additional AGCM experiments for which soil moisture anomalies are initialized as the extreme states from the control simulation. It is found that the relationship between the soil moisture and the near‐surface atmospheric anomalies in these additional experiments is similar to those obtained in the control simulation. This leads us to the conclusions that (1) there is a certain potential for seasonal predictability of the atmospheric surface climate anomalies due to interannual variations in the soil moisture, and (2) the effect of the soil moisture anomalies, however, is mostly confined to the near‐surface climate variability with little impact on the dynamic variability in the atmosphere, for example, in the upper troposphere. We conclude that atmospheric potential predictability associated with soil moisture anomalies is modest and confined to summer months. Any realization of this predictability depends on correctly observing and initializing large‐scale soil moisture anomalies. These conclusions are based on a particular AGCM and its land surface parameterization scheme. More thorough investigations with more sophisticated land surface schemes are certainly warranted.

Future changes in the sign and intensity of ocean–land–atmosphere interactions have been insufficiently studied, despite implications for regional climate change projections, extreme event statistics, and seasonal climate predictability. In response to this deficiency, the present study focuses on projected responses to the enhanced greenhouse effect in: (1) the mean state of the atmosphere and land surface; (2) oceanic and terrestrial drivers of sub-Saharan climate variability; and (3) total seasonal climate predictability of sub-Saharan Africa, a region known for its pronounced land–atmosphere coupling. Analysis focuses on output from 23 Earth System Models in the Coupled Model Intercomparison Project Phase Five for the late twentieth and twenty-first centuries. It is projected that the greatest warming across sub-Saharan Africa will occur over the Sahel, the monsoon season will become more persistent into late summer and autumn, short rains in the Horn of Africa (HOA) will intensify, and leaf area index will increase across the HOA. Stepwise Generalized Equilibrium Feedback Assessment, i.e. a multivariate statistical approach, is applied to the model output over sub-Saharan Africa in order to explore the oceanic and terrestrial drivers of regional climate. The models indicate that the study region’s climate variability is dominated by oceanic drivers, with secondary contributions from soil moisture and very modest impacts from vegetation. Overall, the general model consensus of future projections indicates a concerning diminished seasonal predictability of sub-Saharan African regional climate based on key oceanic and terrestrial predictors and an elevated role of the land surface (associated with soil moisture anomalies) compared to oceanic drivers in regulating regional climate variability.

To adapt forest ecosystems and forest management to climate change, it is essential to know which forest regions and which tree species are resilient to climate variability and which ones are possibly affected most by past and anticipated future changes. In this contribution, for the main forest regions of Türkiye and six tree species, recent climate variability and trends were quantified and statistically correlated to record tree defoliation and vitality. Climate variables considered are maximum temperature (Tmax), minimum temperature (Tmin), mean temperature (Tmean), and total precipitation (Prcp), which are compared to forest health responses recorded as part of the International Cooperative Programme on Assessment and Monitoring of Air Pollution Effects on Forests (ICP Forests) on 277 plots across forests along the Black Sea and Mediterranean regions. In addition, long-term data on satellite measurements of the normalized difference vegetation index (NDVI) were extracted for the same 277 plots for the period 2008–2020. Firstly, 30 years (1991–2020) of reanalysis of climate variables from ECMWF were extracted for all plots; secondly, individual correlations and cross-correlations of climate variables and tree health and vitality were computed for the period 2008–2020 (significance level of 95%) for the four most dominant species from the Black Sea forests (F. orientalis, Q. cerris, P. sylvestris, P. orientalis) and two species from Mediterranean forests (P. brutia and C. libani). Temperature showed a stronger effect on most species than precipitation. Finally, time-lagged correlations were analyzed for seven-time lags (significance level of 95%) to evaluate legacy effect. The analysis revealed that different tree species from the two regions show different responses to climate variables. Species in the Mediterranean region are more resistant to droughts and climatic variations. Legacy effects of defoliation and NDVI have lasted for at least 2 years.

ABSTRACTThe timescale of decadal climate predictions, from a year‐ahead up to a decade, is an important planning horizon for stakeholders in the energy sector. With power systems transitioning towards a greater share of renewable energy sources, these systems become more sensitive to the variability of weather and climate, thus necessitating the provision of long‐range climate predictions to ensure effective planning and operation. As decadal predictions sample both the internal variability of the climate and the externally forced response, these forecasts potentially provide useful information for the upcoming decade. Here, we show for the first time that it is possible to make skillful decadal predictions for a range of energy sector relevant climate variables over the European region. We apply post‐processing techniques and identify skill in certain regions during both summer and winter for temperature, solar irradiance, and precipitation. We also show significant skill for 850 hPa zonal wind speed and the North Atlantic Oscillation during the extended winter period (October–March). We demonstrate how these forecasts can be used for important energy indicators, such as offshore wind capacity factors, comparing the skill of direct model output (using forecast variables directly) and pattern‐based approaches (e.g., using the NAO index). We find significant skill for predictions of modeled European energy variables, including Northern European offshore wind capacity factors (r = 0.73), UK electricity demand (r = 0.84), solar photovoltaic capacity factors in Spain (r = 0.63), and precipitation in Scandinavia (r = 0.64). Our results highlight the potential for skilful prediction of energy‐sector relevant quantities on decadal timescales. This could benefit both the planning and operation of the future energy system.

Green plants play an important role in matter and energy transformations and are key messengers in the carbon and energy cycle. Vegetation net primary productivity (NPP) reflects the capability of plants to convert solar energy into photosynthate (fixed carbon). It can be used as a direct indicator of current status and future trend of ecological processes affected by human activities and climate variability, especially in fragile ecological environments such as karst areas where rock-desertification increasingly occurs. Using remote sensing and geographic information system technologies, research on spatio-temporal variations of NPP and its response to changes in climate variables (temperature, rainfall and solar radiation) in Guizhou province of China was conducted. Karst areas within the study area were delineated to explore the relationship between NPP and climate variables. The NPP model developed by Wang et al. (2010) was used. This model is a modification of the light use efficiency (LUE) model of Running et al. (2000), validated for this region in China. As input to the model, three climate scenarios were designed - one representing current conditions and two for hypothetical future conditions. Both future scenarios have a 2˚C increase in temperature and 6% increase in rainfall, applied spatially across the region using a constant scaling factor on the current condition scenario. The first future scenario has unchanged solar radiation; and the second has a 10% increase in solar radiation for investigating the effect of solar radiation on NPP. The model was run on a monthly time step for one year. Under the current condition scenario, the spatio- temporal patterns of monthly NPP were analysed using MODIS imagery, with the climate variables as model inputs. The relationships between NPP and temperature, and NPP and rainfall, were quantified using single and partial correlation. The hypothetical future condition scenarios were then used to explore the response of NPP to changes in these variables. This paper reports on the results of this modelling under all scenarios, with particular interest in the relationship between solar radiation and karst areas. Model results show that, under the current condition scenario, the mean annual NPP of terrestrial vegetation is 421gCm -2 with the NPP in karst areas being 11.9% lower than in non-karst areas. Maximum and minimum NPP values occur in July and January, respectively. The results show strong correlations between NPP and climatic variables: (1) Temperature is a key factor which significantly limits vegetation growth in the northwest. The correlation between NPP and temperature decreases from north to south, and is stronger in karst (than non-karst) areas. (2) The correlation between NPP and rainfall is most significant in the southeast and west in comparison with other areas, and is lower in karst (than non-karst) areas. Under the first future scenario, the total annual NPP decreases from 74×10 6 tC to 73×10 6 tC; whereas under the second future scenario, annual NPP increases from 74×10 6 tC to 79×10 6 tC. The results from both future scenarios indicate that the largest variation in NPP occurs in karst areas with severe rock-desertification, indicating solar radiation is a dominant factor affecting NPP in karst areas. There are obvious regional differences on spatio-temporal patterns of NPP between karst areas and non-karst areas. The analysis of the interactions among three climate variables - temperature, rainfall and solar radiation has contributed to our understanding of their contributions to NPP, especially in karst areas.

Rainfed maize yield response to management and climate covariability at large spatial scales

In-depth knowledge of smallholder farmers’ perception of changing climate variables such as recurrent and protracted droughts, late onset of rainfall, early cessation of rainfall and their coping adaptation strategies are very significant in designing climate resilient agriculture among smallholder food crop farmers in sub-Saharan Africa (SSA). This paper examines smallholder farmers’ perceptions of climate variability vis-á-vis meteorological and satellite remote sensing data and their implications for climate smart agriculture technologies. Integration of meteorological, satellite remote sensing and farm-level data were used. Multistage sampling procedure was used to select four towns, eight communities and 398 smallholder food crop farmers. Spearmans’ rank correlation coefficient and Standardized Precipitation Index were used to assess the distribution of climate variables. In addition, three vegetation drought characteristic indices, Normalized Difference Vegetation Index (NDVI), Vegetation Condition Index (VCI) and Water Supply Vegetation Index (WSVI) were used to examine drought conditions within the basin. The results indicated that smallholder farmers in the Offin river basin perceived recurrent and prolonged droughts, rising temperatures, late onset of rainfall, early cessation of rainfall, increasing dry spells, reduction in the length of rainfall season and shorten cropping season as a main indicators of climate variability. The findings further revealed that farmers’ perceptions on climate variability strongly agrees with meteorological and satellite remote sensing data which not only demonstrated rising temperature and frequent and prolonged droughts but also late onset and early cessation of rainfall and reduction in growing season rainfall. Smallholder food crop farmers in the Offin river basin have a high awareness of variation in climate condition and have taken coping strategies to reduce the effects of climate change and climate variability. Smallholder food crop farmers in the basin have also adopted climate smart agriculture technologies such as crop management techniques, integrated soil and nutrient management practices, tillage and residue management, small scale irrigation systems, inland valleys cropping and renewable energy systems to increase agricultural productivity and build resilience to climate variability. The policy implication is that, smallholder food crop farmers’ knowledge on climate variability should be considered as a practical input in designing and planning climate variability coping adaptation and mitigation strategies.

<p>The Tree Ring Width (TRW) records are one of the main paleoclimate proxies that estimate the past climate variability. TRW measurements pave the way for scientists to produce sequences from various kinds of trees and reconstruct climate variables over the past years. Understanding the relation between TRW and climate variables in the past would help us analyze climate change events. This study has applied multi-gridded datasets to find the relations and model TRW data with different climate variables in South Korea's northeast. We utilized TRW data related to our case study that is available on the NOAA website; furthermore, we have checked three primary gauges, namely Agmerra (The Modern-Era Retrospective Analysis for Research and Applications), CRU TS4.03 (Climatic Research Unit Time-Series version 4.03), and APHRODITE's (Asian Precipitation-Highly-Resolved Observational Data Integration Towards Evaluation) for climate variables. In the first step, we have checked the relation between the gauges' precipitation data and observation TRW. According to the obtained efficiency criteria, CRU performed the best consequences. In the second step, we have tried to model observation TRW as a dependent variable and four climate variables of CRU (precipitation, minimum temperature, potential evapotranspiration, and diurnal temperature range) as independent ones over 1969-1998. We have created a linear regression model and determined the accurate coefficients for each climate variable. Besides, we have examined the observation TRW and modeled TRW data. The results showed that with <em>R<sup>2 </sup></em><sup> </sup>≈ 0.40 and a <em>p-value</em> of 0.0323, the regression line was linearly significant at the 95 percent significance level. It represents that our model is acceptable. We will extend our model with Artificial Intelligence methods and try to apply other TRW stations in the future step. In this way, we may produce highly accurate models and fill the gaps for future researches.</p>

Decadal climate prediction can provide invaluable information for decisions made by government agencies and industry. Modes of internal variability of the ocean play an important role in determining the climate on decadal time scales. This study explores the possibility of using self-organizing maps (SOMs) to identify decadal climate variability, measure theoretical decadal predictability, and conduct decadal predictions of internal climate variability within a long control simulation. SOM is applied to an 11-yr running-mean winter sea surface temperature (SST) in the North Pacific and North Atlantic Oceans within the Community Earth System Model 1850 preindustrial simulation to identify patterns of internal variability in SSTs. Transition probability tables are calculated to identify preferred paths through the SOM with time. Results show both persistence and preferred evolutions of SST depending on the initial SST pattern. This method also provides a measure of the predictability of these SST patterns, with the North Atlantic being predictable at longer lead times than the North Pacific. In addition, decadal SST predictions using persistence, a first-order Markov chain, and lagged transition probabilities are conducted. The lagged transition probability predictions have a reemergence of prediction skill around lag 15 for both domains. Although the prediction skill is very low, it does imply that the SOM has the ability to predict some aspects of the internal variability of the system beyond 10 years.

I discuss three aspects of research necessary for investigating possible effects of changes in climatic variability on crop yields. Additional information on changed variability effects is needed to further elucidate uncertainties in our knowledge of possible impacts of climate change on agriculture. First, sensitivity analyses of crop responses to shifting change in variability must be performed. Second, investigations of how climatic variability may change under perturbed (e.g., greenhouse gas-warmed) climate conditions should be undertaken. If one has some confidence in estimates of how variability may change, then a third research task is the formation of climate change scenarios that incorporate changes in climatic variability and their application to crop-climate models to determine crop responses. In this chapter, these research tasks are discussed regarding one climate variable, precipitation. I summarize two research projects that have been undertaken to investigate the sensitivity of the CERES-wheat (Triticum aestivum L.) crop model to changes in climatic variability, on daily to annual time scales, for sites in the central Great Plains. I also provide an example of determining possible changes in daily variability of precipitation through analysis of results from two regional climate model experiments, and then go on to describe an example of forming a climate change scenario that incorporates changes in daily precipitation variability estimated from the regional model runs. The CERES-wheat model exhibits considerable sensitivity to changes in precipitation variability, and variability of yields is positively correlated with variability in precipitation. Consideration of changes on both daily and interannual time scales is important. Estimates of changes in wheat yields simulated from a scenario that includes changes in variability are very different from changes in yield determined from a scenario that considers only mean changes in precipitation.

Assessing the response of crop yield to year-to-year climate variability at the field scale is often done using process-based models and regression techniques. Although powerful, these tools rely on strong assumptions and can lead to substantial prediction errors. In this study, we investigate the use of a flexible machine learning algorithm combining Functional Principal Component Analysis and Random Forest, to relate field scale wheat yield to local daily climate variables. Instead of computing seasonal, monthly or any other arbitrary time-frame climate averages, climate time series are decomposed by Functional Principal Component Analysis into a few data-driven basis functions, called Principal Curves, in order to summarize the dynamic of key climate variables by a limited number of interpretable components. Scores associated to these components are then used as inputs of a Random Forest algorithm for yield prediction and for analysing important factors responsible for yield variability. To evaluate our approach, we use a French national database including wheat yield data as well as climate and management practice data for 298 farm fields from 2011 to 2016 in four main producing regions. Depending on the regions, our approach can explain from 62 % to 81 % of the yield variability when both agronomic and climate variables are included, down to 56–81 % when ignoring agronomic variables and 51–74 % when ignoring climate variables. Based on a year-by-year cross-validation, RMSE ranges from 0.5 t ha−1 to 2.1 t ha−1 in non-extreme years (2012–2015). However, prediction error can reach 3.6 t ha−1 in case of exceptional weather conditions, such as those experienced in 2016 in Northern France. We find that this new approach performs in most cases better than the same machine learning algorithm using the usual time averages of climate variables, without the need to choose an arbitrary time-frame. We then show how important patterns in weather time series can be identified and how their effects on yield can be interpreted using the proposed modelling framework.

Climate variability is noted a critical issue in Nepal. This paper investigates empirically and analytically whether climate variability exists or not in different altitudes, whether its direction moves and what will be its future direction. In this study, we used a time series forecasting model based on the secondary data of hydrology and metrology collected from the Department of Hydrology and Metrology, the Government of Nepal. As a result, this paper found climate variability in the different parts of Nepal where variations of climatic variables (temperature and rainfall precipitation) are found all months of the year. In the estimation, on average temperature rising per annum is by 20 C from 1975 to 2012. However, declining rainfall per annum was noticed. Additionally, the estimated forecasting of temperature variation over the next 40 years would be 60c over. In the case of rainfall forecast, it would be constant and results from drought at high altitude and flood at low altitude. Therefore, climate variability may be a big threat in the next 40 years across the country, of Nepal. The result may be valuable input to the policymakers and farmers for economics as usual.