Index Based Analysis of Climate Change Scenarios

Transcription

1 Politecnico di Milano Scuola di Ingegneria Civile, Ambientale e Territoriale Master of Science in Civil, Environmental and Land Management Engineering Index Based Analysis of Climate Change Scenarios Lake Como catchment case study Supervisor: Prof. Andrea F. Castelletti Assistant Supervisor: Dr. Yu Li Master Graduation Thesis by: Sui Xin Student Id n Academic Year

2

3 Attitude, basic skill and psychological quality Lang Ping

4 ACKNOWLEDGMENTS Firstly, I am thankful for my professor Andrea F. Castelletti gave me this chance to join in my favorite project, which I should say, I feel like to join from the beginning when I consider about my final thesis topic. And thanks my professor Andrea gave me an opportunity to meet my Co-supervisor Yu Li. Yu is a very helpful person to give me confidence to conquer every big difficulties during my thesis. Then I m really thankful for Politecnico di Milano, my mother campus. She constructs my logical way of thought and grows me to be calm enough when I encounter every task in life. That is a real value for a human.

5 Table of Contents ACKNOWLEDGMENTS... 4 Abstract... 1 Sommario... 2 PREFACE Introduction BACKGROUND CLIMATE CHANGE SCENARIOS Representative Concentration Pathways (RCPs) Shared Socioeconomic Reference Pathways (SSPs) General Circulation Models (GCMs) MOTIVATION Study area Materials and methods MATERIALS Observational dataset Climate projection dataset METHODS Climate extreme Index Results TRENDS IN ANNUAL TEMPERATURE INDICES TRENDS IN ANNUAL PRECIPITATION INDICES Precipitation in annual precipitation indices Trends in seasonal precipitation indices Discussion Conclusion List of Figures List of Tables Acronyms Bibliography... 47

6

7 ABSTRACT Recent climate extreme events demonstrate the vulnerability of European society to climate-related natural hazards, and there is a strong evidence that climate change will worsen these events in the coming years. In other word, future climate extremes may be very different from today and difficult to predict. To this end, this thesis conducted an in-depth analysis of future climate scenarios with a set of state-of-the-art indices. The analysis is based on the projected climate change time-series under Representative Concentration Pathways (RCP) 2.6 scenario, generated from a number of Global Climate Models (GCMs). The dataset is based on daily max and min temperature, as well as daily precipitation. Lake Como catchment (Italy) is used as the study area to investigate projected changes of the climatic conditions till the end of this century. 27 key indices are computed to depict the future extreme situations (such as extreme dry periods or heavy precipitation), based on which the impacts of climate change to the human society will be interpreted. Those results are also useful for different stakeholders, such as several hydropower companies and agricultural authorities, to identify the critical components from the changing climate conditions and to plan for strategic mitigation measures accordingly. 1

8 SOMMARIO Clima recenti eventi estremi dimostrano la vulnerabilità della società europea per i pericoli naturali connesse al clima, e vi è una forte evidenza che il cambiamento climatico peggiorerà questi eventi nei prossimi anni. In altre parole, futuri eventi climatici estremi possono essere molto diverse da oggi e difficili da prevedere. A tal fine, questa tesi ha condotto un'analisi approfondita degli scenari climatici futuri, con una serie di indice di state-of-the-art. L'analisi si basa sul cambiamento climatico serie temporale proiettata sotto Pathways rappresentativi di concentrazione (RCP) 2.6 scenario, generato da un certo numero di modelli globali climatici (GCM). Il set di dati si basa sulla massima giornaliera e temperature min, cosìcome precipitazioni giornaliere. Lake Como utenza (Italia) viene utilizzato come area di studio per analizzare cambiamenti previsti delle condizioni climatiche fino alla fine di questo secolo. 27 indice chiave sono calcolati per rappresentare le situazioni future estreme (come ad esempio i periodi di siccità estreme o forti precipitazioni), in base ai quali verranno interpretati gli impatti dei cambiamenti climatici per la società umana. Questi risultati sono utili anche per le diverse parti interessate, come le diverse aziende idroelettriche e le autorità agricole, per identificare i componenti critici dalle condizioni climatiche mutevoli e per pianificare misure strategiche di mitigazione di conseguenza. 2

9 PREFACE Being an environmental engineering student for nearly 7 years, I have a really ardently love for the natural and the environment. As the Global Warming becomes more severe, the impact as occurred around my life can be clearly perceived. Before in the winter, snow in Barzio mountain could last for nearly four months. However, in recent years the period is shorten to 3 months, since snow begin to melt in early spring. Consequently, flood problem happens more frequently in autumn in Como city, which makes me wonder how the future will be and what method can we utilize to have a better predict our future. Natural Resource Management subject is my favorite study area during my master period. This subject takes the environmental factors and also humans into account, and aims to support people in better manage the environment in a friendlier and sustainable way. 3

10 1 INTRODUCTION 1.1 BACKGROUND The research on the climate change extremes has progressed enormously, largely due to international coordinated efforts to simulate the future climate projections with collated, quality controlled climate models, which serve as a basis to analyze parameters of interests and associated extreme events that may occur in the future. One such effort has been led by CORDEX project (Giorgi F et al. 2015), who have facilitated the dissemination of climate change dataset over global scale with various parameters available. Further effort has been made through the provision of free standardized software for analysis of extreme events, and through the organization of regional workshops to fill in data gap in data-spare regions (Peterson and Manton, 2008). For the example of such standardized software, just to name a few, CLIMDEX project was set up to produce a suite of in situ and gridded land-based global datasets of indices representing the more extreme aspects of climate. The package contains 27 indices recommended by ETCCDI (Zhang et al. 2011). Another example is the HadDEX dataset, which currently represents the most comprehensive global gridded dataset of temperature and precipitation extremes based daily in situ data available. It has been used in many model evaluation (e.g., Sillmann and Roekner 2008; Alexander and Arblaster, 2009; Rusticucci et al., 2010; Sillmann et al., 2012) and detection and attribution studies (e.g., Min et al. 2011; Morak et al. 2011), in addition to climate variability and trend studies. 1.2 CLIMATE CHANGE SCENARIOS Depicting the plausible picture of changing conditions in the future is critical for understanding the consequence of climate change and for planning the adaptation strategies accordingly. To this end, projection of climate change scenarios via integrated model simulation can serve as a useful basis to facilitate the abovementioned processes. This task has been worked many years by the intergovernmental panel on climate change (IPCC), which is an international body for assessing the science related to climate change. The IPCC was set up in 1988 by the world meteorological organization (WMO) and United Nations environment program (UNEP) to provide policymakers with regular assessments of the scientific basis of climate change, its impacts and future risks, and options for adaptation and mitigation. IPCC assessments provide a scientific basis for governments at all levels to develop climate-related policies, and they underlie negotiations at the UN climate conference - the United Nations Framework Convention on Climate Change (UNFCCC). The assessments are policy-relevant but not policy prescriptive. 4

11 It is worth to mention that one of the main commissions of IPCC is to approve sets of scenarios for climate research, as well as to provide guidance for producing those scenarios. Previous scenario exercises in the climate change research community were developed and applied sequentially in a linear causal chain that extended from the socioeconomic factors that influence greenhouse gas emissions to atmospheric and climate processes to impacts (see Figure 1-1). Detailed socioeconomic scenarios were developed first and used to prepare emissions scenarios, which in turn were used in climate model experiments that formed the basis of climate change projections. Lags in the development process meant that it was often many years until climate and socioeconomic scenarios were available for use in studies of impacts, adaptation, and vulnerability. Basing the scenarios first on socioeconomic characteristics also led to sets that did not necessarily fully span the literature range on future emissions and climate response. In the new process recommended latest fifth report, emissions and socioeconomic scenarios are developed in parallel, building on different trajectories of radiative forcing over time. Rather than starting with detailed socioeconomic storylines to generate emissions and then climate scenarios, the new process begins with a limited number of alternative pathways (trajectories over time) of radiative forcing levels (or CO 2 equivalent concentrations) that are both representative of the emissions scenario literature and span a wide space of resulting greenhouse gas concentrations that lead to clearly distinguishable climate futures. These radiative forcing trajectories were thus termed Representative Concentration Pathways (RCPs). The RCPs are not associated with unique socioeconomic assumptions or emissions scenarios but can result from different combinations of economic, technological, demographic, policy, and institutional futures. In the preparatory phase, each RCP was simulated in an Integrated Assessment model to provide one internally consistent plausible pathway of emissions and land use change that leads to the specific radiative forcing target. The full set of RCPs spans the complete range of integrated assessment literature on emissions pathways and the radiative forcing targets are distinct enough to result in clearly different climate signals. 5

12 Figure 1-1. Approaches to the development of global scenarios: (a) previous sequential approach ; (b) proposed parallel approach. Numbers indicate analytical steps (2a and 2b proceed concurrently). Arrows indicate transfers of information (solid) selection of RCPs Representative Concentration Pathways (RCPs) RCP are four greenhouse gas concentration trajectories adopted by IPCC for its fifth assessment report in It supersedes Special Report on Emissions Scenarios (SRES) projections published in 2000 (Nakicenovic et al. 2000). The pathways are used for climate modeling and research. They describe four possible climate futures, all of which are considered possible depending on how much greenhouse gases are emitted in the years to come. The four RCPs, namely RCP 2.6, RCP 4.5, RCP 6, and RCP 8.5, are named after a possible range of radiative forcing values in the year 2100 relative to pre-industrial values (i.e., + 2.6, + 4.5, + 6.0, and W/m 2, respectively; See Table 1-1). 6

13 Table 1-1. Overview of four RCPs Description RCP 2.6 Rising radiative forcing pathway leading to 2.6 W/m² in RCP 4.5 Stabilization without overshoot pathway to 4.5 W/m² at stabilization after 2100 RCP 6 Stabilization without overshoot pathway to 6W/m 2 at stabilization after 2100 RCP 8.5 Peak in radiative forcing at 8.5 W/m² before 2100 and decline The RCPs are consistent with a wide range of possible changes in future anthropogenic (i.e., human) GHG emissions. RCP2.6 assumes that global annual GHG emissions (measured in CO 2-equivalents) peak between , with emissions declining substantially thereafter. Emissions in RCP4.5 peak around 2040, then decline. In RCP6, emissions peak around 2080, then decline. In RCP8,5, emissions continue to rise throughout the 21st century. The population and GDP pathways underlying the four RCPs are shown in Figure 1-2. It also shows, as reference, the UN population projections and the 90th percentile range of GDP scenarios in the literature on greenhouse gas emission scenarios. It should be noted that, with one exception (RCP8.5), the modeling teams deliberately made intermediate assumptions about the main driving forces (as illustrated by their position in Figure 1-2) (see the relevant papers elsewhere in this Special Issue). In contrast, the RCP8.5 was based on a revised version of the SRES A2 scenario; here, the storyline emphasizes high population growth and lower incomes in developing countries. 7

14 Figure 1-2. Population and GDP projections of the four scenarios underlying the RCPs. Grey area for population indicates the range of the UN scenarios (low and high) (UN 2003). Grey area for income indicates the 98th and 90th percentiles (light/dark grey) of the IPCC AR4 database (Hanaoka et al. 2006). The dotted lines indicate four of the SRES marker scenarios. For energy use, the scenarios underlying the RCPs are consistent with the literature - with the RCP2.6, RCP4.5 and RCP6 again being representative of intermediate scenarios in the literature (resulting in a primary energy use of 750 to 900 EJ in 2100, or about double the level of today; see Figure 1-3). The RCP8.5, in contrast, is a highly energy-intensive scenario as a result of high population growth and a lower rate of technology development. In terms of the mix of energy carriers, there is a clear distinction across the RCPs given the influence of the climate target (for details, see the papers elsewhere in this Special Issue). Total fossil-fuel use basically follows the radiative forcing level of the scenarios; however, due to the use of carbon capture and storage (CCS) technologies (in particular in the power sector), all scenarios, by 2100, still use a greater amount of coal and/or natural gas than in the year The use of oil stays fairly constant in most scenarios, but declines in the RCP2.6 (as a result of depletion and climate policy). The use of non-fossil fuels increases in all scenarios, especially renewable resources (e.g. wind, solar), bio-energy and nuclear power. The main driving forces are increasing energy demand, rising fossil-fuel prices and climate policy. An important element of the RCP2.6 is the use of bio-energy and CCS, resulting in negative emissions (and allowing some fossil fuel without CCS by the end of the century). 8

15 Figure 1-3. Development of primary energy consumption (direct equivalent) and oil consumption for the different RCPs. The grey area indicates the 98th and 90th percentiles (light/dark grey) (AR4 database (Hanaoka et al. 2006) and more recent literature (Clarke et al. 2010; Edenhofer et al. 2010)). The dotted lines indicate four of the SRES marker scenarios Shared Socioeconomic Reference Pathways (SSPs). The SSPs define the state of human and natural societies at a macro scale and have two elements: a narrative story line and a set of quantified measures that define the high-level state of society as it evolves over the 21-st century under the assumption of no significant climate feedback on the SSP. This assumption allows the SSP to be formulated independently of a climate change projection. In reality, SSPs may be affected by climate change, which can be taken into account when combining SSPs with climate change projections to generate a socioeconomic-climate reference scenario. In the absence of climate policies, the SSPs may lead to different climate forcing in the reference case and to different changes in climate. In Figure 1-4 representation, two axes of the scenario matrix are the SSPs and radiative forcing levels. Each combination of an SSP and a radiative forcing level defines a family of macro-scale scenarios. Because the RCP level provides only a rudimentary specification of mitigation policy characteristics, and very little information on adaptation policies, a third axis embeds RCPs in Shared Climate Policy Assumptions (SPAs) that include additional information on mitigation and adaptation policies, e.g. global and sectoral coverage of greenhouse gas reduction regimes, and the aggressiveness of adaptation in different world regions. Obviously, there can be more than one SPA for a given radiative forcing level. For any combination of SSP, RCP, and SPA, there will be a number of possible climate change projections that are associated with a different model of the physical climate system, adding another dimension to each cell. 9

16 Figure 1-4. The scenario matrix architecture: confronting different future levels of climate forcing with different socio-economic reference assumptions described by SSPs General Circulation Models (GCMs) GCMs, which represent physical processes in the atmosphere, ocean, cryosphere and land surface, are the most advanced tools currently available for simulating the response of global climate system to increasing greenhouse gas concentrations. While simpler models have also been used to provide globally or regionally averaged estimates of the climate response, only GCMs, possibly in conjunction with nested regional models, have the potential to provide geographically and physically consistent estimates of regional climate change which are required in impact analysis. The results from GCMs are a set of global dataset that describes future changes of climate forcing under different socio-economic and RCP scenarios. For example, Figure 1-5 shows projected changes worldwide on a regional level in response to different scenarios of increasing carbon dioxide simulated by 21 climate models, and are used to help scientists and planners conduct climate risk assessments to better understand local and global effects of hazards, such as severe drought, floods, heat waves and losses in agriculture productivity. 10

17 Figure 1-5. Coupled Model Inter comparison Project Phase 5 (CMIP5) multi-model mean projections(i.e., the average of the model projections available) for the period under the RCP2.6 (left) and RCP8.5 (right) scenarios for (a) change in annual mean surface temperature and (b) change in annual mean precipitation, in percentages, and (c) change in average sea level. Changes are shown relative to the period. The number of CMIP5 models used to calculate the multi-model mean is indicated in the upper right corner of each panel. Stippling (dots) on (a) and (b) indicates regions where the projected change is large compared to natural internal variability (i.e., greater than two standard deviations of internal variability in 20-year means) and where 90% of the models agree on the sign of change. Hatching (diagonal lines) on (a) and (b) shows regions where the projected change is less than one standard deviation of natural internal variability in 20-year means Yet, in using those dataset one has to tack with spatial resolution of the output from GCMs, which are often too coarse relative to the scale of exposure units in most impact assessment, hence as those processes related to smaller scales might not be properly captured. Instead, their known properties must be averaged over the larger scale in a technique known as parameterization. Other alternatives are also available to mitigate this issue by using downscaling approaches via regional modeling or statistical methods (Sylwia Trzaska, 2014). 11

18 1.3 MOTIVATION The purpose of the current study is to analyze the projected climate change time series using different indices, which may allow to uncover the multi-dimensional facets from climate change dataset. This should allow us to explore the potential stressors beneath the general trend of increasing temperature as reported by many studies. 12

19 2 STUDY AREA Figure 2-1. Map of Lombardy region (Italy). The study area, i.e., Lake Como catchment, is colored in violet located in northern part. Lake Como is the third largest Italian lake located in northern Italy close to Switzerland (see Figure 2-1), and receives water from a catchment of around 4,500 km 2 characterized by a highly varying terrain elevation, which provides a huge hydro-power potential exploited through a series of small to medium artificial reservoirs for a total storage capacity of 545 Mm 3 (green triangles in Figure 2-1). Since 1946, Lake Como is regulated by Consorzio dell Adda1, a consortium of downstream stakeholders mostly composed by farmers but also hydro-power companies and industries. Lake Como has an active storage of around 250 Mm 3 13

20 regulated through a dam on the effluent Adda river and is operated for multiple purposes, including a number of run-of-river hydro-electric power plants and several large agricultural districts (green area downstream shown in Figure 2-1). Besides water supply, the regulation of the lake aims urban flood protection of the lake shores, particularly in Como City. The hydrological regime is influenced by both spring snowmelt and precipitation, resulting into a bi-modal peak (see Figure 2-2 green and blue plots): one more pronounced peak corresponding to the snow-melt season, in late spring, and a smaller but more variable one produced by autumn rains. The spring peak (from May to July) is the most important contribution to the seasonal storage, which is released during the summer, when the agricultural water demand is high (violet plot in Figure 2-2). While water availability has not been an issue for many years, in recent decade more drought events have been reported. Figure 2-3 visualizes the trend in the inflows observed over the last 60 years using a tool called Moving Average over Shifting Horizon (MASH; Anghileri, Pianosi, and Soncini-Sessa 2014), a trend analysis technique that aims to identify non-stationary changes in hydro-climatic variables. As shown in Figure 2-3, there is a clear decreasing trend of inflow during the late spring and summer periods, which are the most critical for irrigated agriculture. In fact, in 2003, 2005, 2006 and 2007 the system experienced severe drought events, which caused great losses to the agriculture (Anghileri, Pianosi, and Soncini-Sessa, 2014). If this tendency continues over next years, the system is likely to lose its designed functionality, and adoption of adaptation strategies will be indispensable. 14

21 Figure 2-2. Hydrological features of Lake Como estimated as the mean of statistics between , and the nominal water demand trajectory is given by historical regulation policy. Notice that the natural storage estimates are obtained by regression method assuming no regulation imposed. 15

22 Figure 2-3. Trend analysis of the daily inflows over the time horizon , with colors of each line representing the average extra-annual inflow series from every 20-year moving average estimates. The average daily inflow is computed by means of a moving window that includes data over consecutive days in the same year and over the same days in consecutive years, with the horizon of consecutive years progressively shifted ahead to identify long-term trends. Recent hydrological extreme events demonstrate the vulnerability of European society to water-related natural hazards, and there is strong evidence that climate change will worsen these events in the coming years. Future hydrological extremes may be very different from today s reality and difficult to predict. Changed hydro-climatic extremes will have important implications on the water sector and the design of water management practices. Therefore, there is an urgent need to better understand the changing conditions projected in the future, and the potential stressors that associate with such changing conditions. 16

23 3 MATERIALS AND METHODS 3.1 MATERIALS Observational dataset E-OBS gridded version of the ECA dataset with daily temperature and precipitation is used as ground truth observation. The ECA dataset contains series of daily observations at meteorological stations throughout Europe and the Mediterranean. The data files contain gridded data for 4 variables (daily mean temperature, daily minimum temperature, daily maximum temperature and daily total precipitation). Table 3-1 summarizes the main feature of the EOBS dataset, and Figure 3-1 shows the domain of our study area within the EOBS grids. Table 3-1. Feature of EOBS observatory dataset Description Time span 1/1/ /31/2015 Spatial resolution Temporal resolution 0.25 by 0.25 degree daily Figure 3-1. Map of EOBS grid dataset for the study area 17

24 3.1.2 Climate projection dataset The World Climate Research Program (WCRP) established in 2009 the Task Force for Regional Climate Downscaling (TFRCD), Which created the CORDEX initiative to generate regional climate change projections for all terrestrial regions of the globe within the timeline of the Fifth Assessment Report (AR5) and beyond. The major aims of the CORDEX initiative are to provide a coordinated model evaluation framework, a climate projection framework, and an interface to the applicants of the climate simulations in climate change impact, adaption, and mitigation studies (Giorgi et al., 2009). The EURO-CORDEX is a branch of CORDEX initiative that produces ensemble climate simulations based on multiple dynamical and empirical-statistical downscaling models forced by multiple global climate models from Coupled Model Inter Comparison Project Phase 5 (CMIP5). Simulations of EURO-CORDEX consider the global climate simulations from the CMIP5 long term experiments up to the year They are based on greenhouse gas emission scenarios (i.e., RCPs) corresponding to stabilization of radiative forcing after the 21st century (Moss et al., 2010 and 2008; Nakicenovic et al., 2000; Van Vuuren et al., 2008). In this study, we adopted the climate projection dataset RCP2.6 radiative scenarios, meaning the optimal situation assuming GHG emissions peak between and decline substantially thereafter. Table 1-1 summarizes the main feature of the EOBS dataset, and Figure 3-2 shows the domain of our study area within the EOBS grids. Table 3-2. Features of EURO-CORDEX dataset Description Time span for projection for control period Spatial resolution Temporal resolution 0.44 by 0.44 degree daily 18

25 Figure 3-2. The map of study area within EURO-CORDEX grid domain. 3.2 METHODS Climate extreme Index To monitor changes in climate and climate extremes, a set of key indices has been computed (e.g. Frich et al., 2001). A good index is expected to have a clear meaning, be highly relevant to people, provide insights into climate change, be homogeneous, easy to understand, be relevant to the practical concerns of policy makers and should not smooth out potentially important changes. Guided by those criteria, a set of 27 indices indicative of climate change are structured based on CLIMDEX project. A detail definition on each index is provided below: Those index decompose the projection of climate change time-series into various aspects representing the extreme events. Indices are derived from daily temperature and precipitation data using the definitions recommended by the expert team on Climate Change Detection and Indices (ETCCDI). 1. FD. Number of frost days: Annual count of days when TN (daily minimum temperature) < 0. Let TNij be daily minimum temperature on day i in year j. Count the number of days where: TNij < 0. 19

26 2. SU. Number of summer days: Annual count of days when TX (daily maximum temperature) > 25. Let TXij be daily maximum temperature on day i in year j. Count the number of days where: TXij > ID. Number of icing days: Annual count of days when TX (daily maximum temperature) < 0. Let TXij be daily maximum temperature on day i in year j. Count the number of days where: TXij < TR. Number of tropical nights: Annual count of days when TN (daily minimum temperature) > 20. Let TNijbe daily minimum temperature on day i in year j. Count the number of days where: TNij > GSL. Growing season length: Annual (1st Jan to 31st Dec in Northern Hemisphere (NH), 1st July to 30th June in Southern Hemisphere (SH)) count between first span of at least 6 days with daily mean temperature TG>5 and first span after July 1st (Jan 1st in SH) of 6 days with TG<5. Let TGij be daily mean temperature on day i in year j. Count the number of days between the first occurrence of at least 6 consecutive days with: TGij > 5. And the first occurrence after 1st July (1st Jan. in SH) of at least 6 consecutive days with: TGij < TXx. Monthly maximum value of daily maximum temperature: Let TXx be the daily maximum temperatures in month k, period j. The maximum daily maximum temperature each month is then: TXxkj=max(TXxkj) 7. TNx. Monthly maximum value of daily minimum temperature: Let TNx be the daily minimum temperatures in month k, period j. The maximum daily minimum temperature each month is then: TNxkj=max(TNxkj) 8. TXn. Monthly minimum value of daily maximum temperature: Let TXn be the daily maximum temperatures in month k, period j. The minimum daily maximum temperature each month is then: TXnkj=min(TXnkj) 9. TNn. Monthly minimum value of daily minimum temperature: Let TNn be the daily minimum temperatures in month k, period j. The minimum daily minimum temperature each month is then: TNnkj=min(TNnkj) 10. TN10p. Percentage of days when TN < 10th percentile: Let TNij be the daily minimum temperature on day i in period j and let TNin10 be the calendar day 10th percentile centered on a 5-day window for the base period The percentage of time for the base period is determined where: TNij < TNin10. To avoid possible inhomogeneity across the in-base and out-base periods, the calculation for the base period ( ) requires the use of a bootstrap processes. Details are described in Zhang et al. (2005). 11. TX10p. Percentage of days when TX < 10th percentile: Let TXij be the daily maximum temperature on day i in period j and let TXin10 be the calendar day 10th percentile centered on a 5-day window for the base period The percentage of time for the base period is determined where: TXij < TXin10.To avoid possible inhomogeneity across the in-base and out-base 20

27 periods, the calculation for the base period ( ) requires the use of a bootstrap processes. Details are described in Zhang et al. (2005). 12. TN90p. Percentage of days when TN > 90th percentile: Let TNij be the daily minimum temperature on day i in period j and let TNin90 be the calendar day 90th percentile centered on a 5-day window for the base period The percentage of time for the base period is determined where: TNij > TNin90. To avoid possible inhomogeneity across the in-base and out-base periods, the calculation for the base period ( ) requires the use of a bootstrap procedure. Details are described in Zhang et al. (2005). 13. TX90p. Percentage of days when TX > 90th percentile: Let TXij be the daily maximum temperature on day i in period j and let TXin90 be the calendar day 90th percentile centered on a 5-day window for the base period The percentage of time for the base period is determined where: TXij > TXin90. To avoid possible inhomogeneity across the in-base and out-base periods, the calculation for the base period ( ) requires the use of a bootstrap procedure. Details are described in Zhang et al. (2005). 14. WSDI. Warm spell duration index: Annual count of days with at least 6 consecutive days when TX > 90th percentile. Let TXij be the daily maximum temperature on day i in period j and let TXin90 be the calendar day 90th percentile centered on a 5-day window for the base period Then the number of days per period is summed where, in intervals of at least 6 consecutive days: TXij > TXin CSDI. Cold spell duration index: Annual count of days with at least 6 consecutive days when TN < 10th percentile. Let TNij be the daily maximum temperature on day i in period j and let TNin10 be the calendar day 10th percentile centered on a 5-day window for the base period Then the number of days per period is summed where, in intervals of at least 6 consecutive days: TNij < TNin DTR. Daily temperature range: Monthly mean difference between TX and TN. Let TXij and TNij be the daily maximum and minimum temperature respectively on day i in period j. If I represent the number of days in j, then: DTR I i 1 j ( Tx ij I Tn ) 17. Rx1day. Monthly maximum 1-day precipitation: Let RRij be the daily precipitation amount on day i in period j. The maximum 1-day value for period j are: Rx1dayj = max (RRij) 18. Rx5day. Monthly maximum consecutive 5-day precipitation: Let RRkj be the precipitation amount for the 5-day interval ending k, period j. Then maximum 5-day values for period j are: Rx5dayj = max (RRkj) ij 21

28 19. SDII. Simple precipitation intensity index: Let RRwj be the daily precipitation amount on wet days, w (RR 1mm) in period j. If W represents number of wet days in j, then: SDII W w 1 j ( RR 20. R10mm. Annual count of days when PRCP 10mm: Let RRij be the daily precipitation amount on day i in period j. Count the number of days where: RRij 10mm 21. R20mm. Annual count of days when PRCP 20mm: Let RRij be the daily precipitation amount on day i in period j. Count the number of days where: RRij 20mm 22. Rnnmm. Annual count of days when PRCP nnmm, nn is a user defined threshold: Let RRij be the daily precipitation amount on day i in period j. Count the number of days where: RRij nnmm 23. CDD. Maximum length of dry spell, maximum number of consecutive days with RR < 1mm: Let RRij be the daily precipitation amount on day iin period j. Count the largest number of consecutive days where: RRij < 1mm 24. CWD. Maximum length of wet spell, maximum number of consecutive days with RR 1mm: Let RRij be the daily precipitation amount on day iin period j. Count the largest number of consecutive days where: RRij 1mm 25. R95pTOT. Annual total PRCP when RR > 95p. Let RRwj be the daily precipitation amount on a wet day w (RR 1.0mm) in period i and letrrwn95 be the 95th percentile of precipitation on wet days in the period. If W represents the number of wet days in the period, then: W R95 P j RR wj where RR RRwj RRwn95 w R99pTOT. Annual total PRCP when RR > 99p: Let RRwj be the daily precipitation amount on a wet day w (RR 1.0mm) in period i and letrrwn99 be the 99th percentile of precipitation on wet days in the period. If W represents the number of wet days in the period, then: W R99P j RR wj where RR RRwj RRwn99 w PRCPTOT. Annual total precipitation in wet days: Let RRij be the daily precipitation amount on day i in period j. If I represents the number of days in j, then PRCPTOT j W wj ) I RR ij i 1 22

29 Those indices can be divided into 5 different categories: 1. Percentile-based indices include occurrence of cold nights (TN10p), occurrence of warm nights (TN90p), occurrence of cold days (TX10p), occurrence of warm days (TX90p), very wet days (R95p) and extremely wet days (R99p). The temperature percentile-based indices sample the coldest and warmest deciles for both maximum and minimum temperatures, enabling us to evaluate the extent to which extremes are changing. The precipitation indices in this category represent the amount of rainfall falling above the 95th (R95p) and 99th (R99p) percentiles and include, but are not be limited to, the most extreme precipitation events in a year. 2. Absolute indices represent maximum or minimum values within a season or year. They include maximum daily maximum temperature (TXx), maximum daily mini-mum temperature (TNx), minimum daily maximum temperature (TXn), minimum daily minimum temperature (TNn), maximum 1-day precipitation amount (RX1day) and maximum 5-day precipitation amount (RX5day). 3. Threshold indices are defined as the number of days on which a temperature or precipitation value falls above or below a fixed threshold, including annual occurrence of frost days (FD), annual occurrence of ice days (ID), annual occurrence of summer days (SU), annual occurrence of tropical nights (TR), number of heavy precipitation days > 10 mm (R10) and number of very heavy precipitation days > 20 mm (R20). These indices are not necessarily meaningful for all climates because the fixed thresholds used in the definitions may not be applicable everywhere on the globe. However, previous studies (e.g., Frich et al., 2002; Kiktev et al., 2003) have shown that temperature indices such as FD, the number of days on which minimum temperature falls below 0LC, have exhibited coherent trends over the mid latitudes during the second half of the 20th century. In addition, changes in these indices can have profound impacts on particular sectors of society or eco-systems. So we included the indices in our study, even though some of them may not provide truly global spatial coverage or be truly extreme. 4. Duration indices define periods of excessive warmth, cold, wetness or dryness or in the case of growing season length, periods of mildness. They include cold spell duration indicator (CSDI), warm spell duration indicator (WSDI), growing season length (GSL), consecutive dry days (CDD) and consecutive wet days (CWD). Many of these indices were used in the near global analysis of Frich et al. (2002). The heat wave duration index (HWDI) defined by Frich et al. (2002) has been found not to be statistically robust as it had a tendency to have too many zero values (Kiktev et al., 2003). This is because Frich et al. (2002) used a fixed threshold to compute the index. This threshold is too high in many regions, such as the tropics, where the variability of daily temperature 23

30 is low. To overcome this, the ETCCDMI replaced this index with the warm spell duration index (WSDI) which is calculated using a percentile based threshold. As this index only sampled daytime maxima we also chose to analyze spells of nighttime minima (CSDI). The CDD index is the length of the longest dry spell in a year while the CWD index is defined as the longest wet spell in a year. This category of indices also includes the length of the growing season (GSL) which is an index that is generally only meaningful in the Northern Hemisphere extra tropics. Other indices include indices of annual precipitation total (PRCPTOT), diurnal temperature range (DTR), simple daily intensity index (SDII), extreme temperature range (ETR) and annual contribution from very wet days (R95pT). They do not fall into any of the above categories but changes in them could have significant societal impacts. All of the climate indices can be calculated using precipitation, maximum temperature, and minimum temperature dataset. Those indices were chosen primarily for assessment of the many aspects of a changing global climate which include changes in intensity, frequency and duration of temperature and precipitation events. They represent events that occur several times per season or year giving them more robust statistical properties than measures of extremes which are far enough into the tails of the distribution so as not to be observed during some years. Together they enabled the presentation of an up-to-date and comprehensive picture of trends in extreme related to temperature and precipitation changes. Before initializing the computation those index, a few steps are required to preprocess the observation and climate change projection dataset. The workflow of preprocessing of required inputs can be described below: 1. Quality control of the data to fill the missing value and to construct a complete time series. 2. Aggregation of spatial distributed dataset over Lake Como basin using Thiessen polygon approach. 3. Bias-correction and downscaling of climate projection data using quantile mapping approach. Once the final dataset is prepared, the 27 climate change indices were calculated. 24

31 4 RESULTS 4.1 TRENDS IN ANNUAL TEMPERATURE INDICES All temperature-related indices show significant and widespread warming trends, which are generally stronger for indices calculated from daily minimum (night time) temperature than for those calculated from daily maximum (daytime) temperature. For example, the frequency of cool nights based on daily minimum temperatures is shown to have significant decreased almost everywhere during this century (see Figure 4-5). The strongest reductions, up to 15 days between 2005 and this century period. Correspondingly, at the upper tail of the minimum temperature of warm nights in almost all seasons (Figure 4-5). Globally average, the frequency of warm nights has increased by about 55% (20 days in a year) during the 100 years time series. 98% show significant increases in TN90p and decreases in TN10p, respectively. Mostly warming trends are also apparent in the absolute warmest and coldest temperature of the year. The warming is generally stronger for the coldest than for the warmest value. Since the middle of the 21st century the coldest night (See Figure 4-16) and coldest day (see Figure 4-14) of the year, for example, have significantly increased over Como region. Warming trends are particularly strong (up to 1 ) over Como lake region. The result also shows significant increases in TNn (TXn) during period, whereas significant decreases are only found in temperature related to the coldest night of year (TNn) has increased by about 4 in the time series. 25

32 Figure 4-1. Change in the number of frost days. Figure 4-2. Change in the number of summer days. 26

33 Figure 4-3. Change in the number of icing days. Figure 4-4. Change in growing season length. 27

34 Figure 4-5. Change in cool nights based on the 10 th percentile of control periods. 28

35 Figure 4-6. Change in cool days. Figure 4-7. Change in the percentage warming nights based on 10 th percentile of control periods. 29

36 Figure 4-8. Change in the percentage of warm days based on 90 th percentile of control periods. Figure 4-9. Change in the number of tropical nights. 30

37 Figure Change in daily temperature range Figure Change in the number of old spell durations. 31

38 Figure Change in warm spell duration. Figure Change in hottest days. 32

39 Figure Change in coldest days. Figure Change in warmest nights. 33

40 Figure Change in coldest night. 4.2 TRENDS IN ANNUAL PRECIPITATION INDICES Precipitation in annual precipitation indices Most of the precipitation indices show (partly significant) changes toward more intense precipitation over Como region. For example, for the number of heavy precipitation days (R10mm and R20mm) and the contribution form very wet days (R95Ptot, See Figure 4-23). Globally averaged, to the indices display upward trends during the 95 years. similar patterns of change are also found for the average intensity of daily precipitation (See Figure 4-21) all precipitation based indices show larger areas with significant trends toward wetter condition than area with drying trends. The number of consecutive dry days (CDD, see Figure 4-19), a measure for extremely dry conditions, also shows trend toward shorter duration of dry spells (i.e., fewer CDD) over Como lake region Trends in seasonal precipitation indices Only two of the precipitation indices, RX1day and RX5day, have data available for sub-annual timescales. We calculated the seasonal values of both indices as the seasonal maxima of the monthly gridded fields. Seasonal trends are generally comparable with annual trends. The annual maximum consecutive 5-day precipitation 34

41 amounts, for example, displays significant tendencies toward stronger extreme precipitation over Como lake region (See in Figure 4-26). In this area, the increase in extreme precipitation is visible across all seasons (See Figure 4-24), but mores significant during winter and autumn. Figure Change in the number of heavy precipitation days. Figure Change in number of extreme precipitation days. 35

42 Figure Change in consecutive dry days. Figure Change in consecutive wet days. 36

43 Figure Change in annual total precipitation. Figure Change in total annual precipitation during wet days. 37

44 Figure Change in fraction of annual total precipitation that exceeds 95th percentile based on control period. Figure Change in the fraction of annual total precipitation that exceeds 99th percentile. 38

45 Figure Change of maximum 1-day precipitation per each month. Figure Change of maximum consecutive 5-day precipitation per each month. 39

46 5.DISCUSSION The individual indices provided insights into recent climate change in Lake Como catchment and some clearly highlighted dramatic changes in the climate of the region as a whole. For example, since 2030, showed strong, nearly linear increases in the number of warm nights (90th percentile of minimum temperature). Apparent increases in the incidence of worldwide annual warm extremes since 2015 mainly reflected the overall warming. However, there has been a larger decrease in cold extremes, leading to a tendency to a decreasing total of all extremes worldwide. This again shows a greater decrease in cold extremes than increase in warm extremes. This may result from changes in regional atmospheric circulation. But in high GHG emission scenario, the situation is more serious. 1. FD is assumed to decrease as a result of a general increase in local and global mean temperature and Su is assumed to increase with the same reason. FD and Su data effect on agriculture, gardening and recreation especially in Como lake region. 2. ID is expected to decrease as a result of warming environment. GSL is expected to increase as a direct result of increasing temperatures and indirectly as a result of reductions in snow cover. GSL is important for agriculture. 3. TN is a direct measure of the number of warm nights. The trend in Figure 4-5 and Figure 4-6 Show apparently increasing trend of warm nights, around 0.1 percentile more than nowadays until This indicator could reflect potential harmful effects of the absence of nocturnal cooling, a main contributor to heat related stress. Summer night-time warming is expected in a greenhouse gas forced climate. This will partly come about as a clear sky radiative effect, partly be a result of increased cloud cover from additional humidity being available for nocturnal condensation. Main effect is expected in late summer, when atmosphere holds maximum amount of moisture. 4. TR, the extraordinarily large number of TN days occurred in 2046, 2081 and The Figure 4-9 shows the large increase in tropical nights. It is important for the human wellbeing that the body can cool down after a hot day. In tropical nights, the temperature stays above 20. During these night, it is more difficult for human body to cool down, especially for elderly or sick people. Therefore, an increase of tropical nights can lead to a rise of mortality. The energy sector is affected by a higher electricity demand during summer due to increased use of air conditioning. 40

47 5. DTR (See Figure 4-10) shows trend for Decrease in the DTR were identified in the Como lake region, where large-area trends show that maximum temperatures have remained constant or have increased only slightly, whereas minimum temperatures have increased at a faster rate. local effects such as urban growth, irrigation, desertification, and variations in local land use can all affect the DTR. 6. TXn, TXn, TNx, TXx are respectively (a) hottest night (TXn) in, (b)coldest day (TXn) in,(c) warmest night (TNx) in and (d) hottest day (TXx) in. Details of trend and time series calculations as described in Figure Figure Warming (but mostly weaker) trends are also found for temperatures related to the warmest night (TNx) and the warmest day (TXx) in Como lake region. On average globally, both Tnx and TXx have increased by about 2 since 2015; over Como lake region, the increases in TNn are stronger than increases in TXx. Consequently, the extreme temperature range is reduced. 7. T90max and T10min are measure of intensity for extreme summer and winter temperature conditions. Both cold days and warm days are in an increasing trend. These are essential to describe timescale appropriate for variability description and societally sensitive extremes (Jones et al. 1999). The change in T90max temperature is higher in winter than summer whereas in the case of T10max the change is higher in summer. 8. From CSDI changes and WSDI changes, the projected warm spell over Como lake region seems to be increasing whereas the cold spells having decreasing trend. 9. Rmm a direct measure of the number of very wet days. This indicator is highly correlated with total annual and seasonal precipitation in most climates. Both of the trends increase obviously in mid of this century, around Greenhouse gas forcing Would lead to a perturbed climate with an enhanced hydrological cycle. More water vapor available for condensation should give rise to a clear increase in the number of days with heavy precipitation extreme precipitation is found. 10. The number of consecutive dry days (CDD) and consecutive wet days (CWD), a measure for extremely dry conditions and wet conditions, also show trends toward shorter duration of dry spells and shorter duration of wet spells (i.e., fewer CDD and fewer CWD) over Como lake region. CDD, CWD and SDII Effect on vegetation and ecosystems potential drought indicator. A decrease trend in Figure 4-19 and Figure 4-20 reflect a wetter climate, due to more frequent wet days. Under sustained greenhouse gas forcing, the interior of continents may experience a general drying due to increased evaporation. 11. RX1day and RX5day are measure of short-term precipitation intensity Potential flood indicator. The annual maximum consecutive 5day precipitation 41