Review of Cavalcante et al., manuscript number egusphere-2024-650

"From insufficient rainfall to livelihoods: understanding the cascade of drought impacts and policy implications"

1. Summary

This research article explores the cascading impacts of drought as a chain reaction across drought-sensitive sectors, and the role of policy responses in mitigating these cascading effects. Focusing on Ceará state in northeast Brazil, the study utilizes a comprehensive dataset on drought impacts and analyzes local public policy documents. By investigating how different policy measures alleviate drought impacts, the authors demonstrate effective strategies for drought management.

2. General comment

I find this work highly relevant for advancing our understanding of how policies shape impacts. One aspect I particularly appreciate is the use of a unique continuously monitored drought impacts dataset, complemented by stakeholder interviews and the review of relevant policy documents. Furthermore, the study sheds light on socio-environmental-economic impacts that are often overlooked. It also delves into the intersection of drought management and policy sciences, providing insights into the role of public policies in mitigating drought impacts. While the text is well-written and the manuscript is generally well-structured and delivers its messages, I believe that the manuscript can be improved by addressing the minor aspects outlined below.

3. Specific comment

1. Introduction

Line 53-65: This paragraph introduces different types of droughts. I wonder why environmental/ecosystem type of drought is not included among meteorological, agricultural and hydrological types, which is still a physical manifestation of impact before introducing the socio-economic drought? (pls refer to AghaKouchak et al. 2021). If the reason for this is the limited number of impact reports for environmental drought (as shown in Wildfires category in S1 Table 3), leading to its merging with socio-ecological-economic impact category, I would suggest first introducing the environmental drought and its physical impacts, and then explaining why they were merged.

Line 114-116: I assume you focused on smallholders, because “they are one of the most vulnerable group to climate extreme impacts” and they are also called “family agriculture”? I would slightly rephrase these 2 sentences to make this clear. Also would replace “group” with “groups”.

2. Methodology

Point 2.1: A map could help a reader to locate Ceara state and visually compare it to other states.

Line 125-126: “The state has various economic activities, mainly the industrial, textile and automotive sectors, and tourism related to its tropical beaches and wind sports".  Aren't textile and automobile industries part of the industrial sectors?

Table 1: I would leave a space before “July 2019, November and December 2021 and April 2022” to make it more clear that it indicates the period of Field work data.

2.2.3. Policy documents data

Line 192-193: "Policy documents were collected to understand the objectives and strategies of relevant policies and programs in the study area". I would advise specifying more what is meant by “relevant”. E.g. policies related to supporting farmers and their families etc.

3. Results

Line 215-216: "These impacts include: (8), high production costs (9), and socioeconomic impacts* (11)". Is there an impact type missing in front of (8)? Should be “Wildfires” based on Fig.2?

Figure 2: I would add “impacts” to socio-env-eco and start with the capital letter “Socio-env-eco”. Could potentially move these titles to the top within each box and place them horizontally.

Figure 3: It's a bit difficult to read the sentence “Socio-environmental-economical impacts of drought” within the last box, would recommend making the contrast more visible. Same for Figure 6 as Figure 3 is part of it.

Figure 3: “illustration illustrating” in the Figure’s caption.

Line 291: "Policy response path C, from insufficient rainfall to agricultural impacts of drought". However, the “C” is currently placed between agricultural and socio-envi-eco impacts in Figure 4? I would even keep the “C” where it is and change the description of this policy response to “from agricultural and socio-envi-eco impacts” because this “Garantia Safra” helps agricultural impacts not to cascade into socio-economical ones since the crop losses already occurred and the policy provides cash transfers to alleviate the economic losses.

Line 308-310: "The data analysis indicates that socio-environmental-economic impacts have the lowest frequency of reporting, suggesting that public policies have been effective in alleviating the cascade of impacts". I would not generalize this conclusion for environmental impacts – the farmers/observers might not report on the state of ecosystems incl forest, freshwater ecosystem, water quality in lakes/rivers etc. Also, there were no policies mentioned that were alleviating specifically environmental impacts.

4. Discussion

Line 413: "On drought related policies, they remain reactive, such as the crop insurance implement after drought impacts are experienced". This presents an opportunity to mention an example of moving from reactive to proactive policies by e.g. using cash transfers based on forecasted impacts rather than responding to those that have already occurred, highlighting the importance of forecasting and associated proactive drought management.

Supplementary

S1. Table 2: A little unclear with the headings: I'm guessing that "Survey Questions", "Alleviating Factor" shouldn't have a circle in front since they're the same level as the rest of the headings?

S1.3 Policy documents data "The selected documents were about the public policies reported by both farmers and observers in the interviews": have you considered checking other relevant documents that were not mentioned by farmers/observers, just generally from legislative repositories? In case there are some policy documents that potentially help alleviate prevailing hydrological impacts, but are not used by farmers for some reason?

References:

AghaKouchak, A., Mirchi, A., Madani, K., Di Baldassarre, G., Nazemi, A., Alborzi, A., et al. (2021). Anthropogenic drought: Definition, challenges, and opportunities. Reviews of Geophysics, 59, e2019RG000683. https://doi. org/10.1029/2019RG000683

Overarching comments

The paper makes use of relevant, state-of-the-art data on drought impacts, including extracting them from diverse sources. Being able to identify and confidently extract impacts from diverse types of narrative data is a strength of this paper. The layered approach of survey responses, interviews and policy documents is innovative and highly appropriate in this context. It is a good example of developing valuable information for program assessment or decision-making in a low-data environment.

One of the most interesting and perhaps most important takeaways from this paper is evidence that programs to bolster farmers’ income help reduce the worst consequences of drought, such as hunger, thirst and migration. The level of detail is helpful, and there is a fairly clear implication that social support goes a long way in alleviating impacts of physical disaster on human well-being.

I suggest that the authors revise the paper to emphasize its contribution for decision-making in low-data environments, and own their methods as a valuable epistemic contribution – and not apologize for not having enough data for machine-learning. The high-data environment of developed countries should not be considered the global standard or necessary for decision-making, particularly given that some of the most drought-vulnerable parts of the world may have less robust media and information ecosystems.

Drought definitions

Given the interconnected, cascading nature of different types of drought and drought impacts, it is perhaps inevitable that questions arise about how many distinctions is too many and how many is enough.

I recommend that the authors add ecological drought to the paragraph on types of drought – lines 53-65 – with a citation to work by Crausbay et al.:

Crausbay, S. D., Ramirez, A. R., Carter, S. L., Cross, M. S., Hall, K. R., Bathke, D. J., ... & Sanford, T. (2017). Defining ecological drought for the twenty-first century. Bulletin of the American Meteorological Society, 98(12), 2543-2550. https://doi.org/10.1175/BAMS-D-16-0292.1

Also, the statement on lines 53-54 isn’t quite right. The different categories of drought relate more to context and discipline, not to characteristics such as duration, extent and intensity. Tsakiris is saying that drought indices that help measure those characteristics can work with each type of drought, not that indices are used to categorize drought.

As first mentioned on line 180, grouping “socio-environmental-economic” impacts of drought is quite a large catch-all category. Also, here you are acknowledging environmental (ecological?) drought but it is grouped with socio-economic drought. That group has quite a span. It is possible that the main uniting feature of socio-economic and environmental impacts is that people are less likely to recognize and describe them. Please consider splitting environmental impacts from socio-economic, or at least providing a good explanation for why they are grouped that way. I’d consider socio-economic vulnerability to be quite distinct from environmental vulnerability. A group this large and diverse increases the risk that you will miss or oversimplify causal pathways.

Counting mentions

Lines 308-310: You can’t conclude that successful policies are responsible for reducing the cascade of impacts based on frequency of mentions. To draw that conclusion you would need quantitative measurements of individual impacts, such as migration or reduced income, not the number of times that those impacts were mentioned in reports. … Or you would need evidence that the low number of mentions of socio-economic impacts isn’t due to the way the data is collected or to what people pay attention to. Typically, tangible impacts such as reduced crop yield are much easier to notice and report than less tangible or less direct impacts.

Lines 351-352: Again, be very careful about imputing any meaning to the quantity of hydro-related reports. It may be more relevant to cite the content, something along the lines of “many reports underscored water supply challenges remaining despite the construction of reservoirs.”

Narrative strength

But the example from the interview starting on line 310 is good. You could make it stronger by more clearly delineating which experiences were from 1993 and which from 2012-2018. One way would be to add “In contrast” to the start of the sentence that begins on line 316.

Discussion

If I am reading this correctly, it suggests that programs such as PAA and PNAE can block the cascade of impacts by giving farmers increased income. This is actually a key point for reducing societal vulnerability to drought – give people enough resources to have options. But it seems as though you may be giving too much weight to counter-arguments on lines 334-343, and/or you could do more to reconcile two conflicting views. I couldn’t tell from this article whether the people arguing that assistance doesn’t solve the underlying problem believe that is true for all forms of cash transfer programs or for specific programs, or whether they are defining the problem in a way that doesn’t separately consider human well-being.

Lines 362-372: This is a good paragraph, pulling out some of the nuance in the data.

Lines 373-382: This paragraph acknowledges some of the unresolved complexity, but please consider deleting it. Instead, lean into the excellent narrative accounts that you have collected. As you note beginning on line 395, you don’t need artificial intelligence to understand what’s happening to people on the ground. You have done a good job of listening to them and giving them a voice.

Line 409: What are natural values?

Miscellaneous

Line 154: On Table 1, the delineation between the period of data collection for drought impacts monitoring data and for field work data is unclear.

Please review capitalization and punctuation around the use of “e.g.”